Merge branch 'main' into devin/1738752192-fix-memory-reset-openai-dependency

* docs: fix long term memory class name in examples

- Replace EnhanceLongTermMemory with LongTermMemory to match actual implementation
- Update code examples to show correct usage
- Fixes #2026

Co-Authored-By: Joe Moura <joao@crewai.com>

* docs: improve memory examples with imports, types and security

- Add proper import statements
- Add type hints for better readability
- Add descriptive comments for each memory type
- Add security considerations section
- Add configuration examples section
- Use environment variables for storage paths

Co-Authored-By: Joe Moura <joao@crewai.com>

* Update memory.mdx

* Update memory.mdx

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Joe Moura <joao@crewai.com>
Co-authored-by: João Moura <joaomdmoura@gmail.com>

README.md

@@ -1,10 +1,18 @@
 <div align="center">
 
-![Logo of CrewAI, two people rowing on a boat](./docs/crewai_logo.png)
+![Logo of CrewAI](./docs/crewai_logo.png)
 
 # **CrewAI**
 
-🤖 **CrewAI**: Production-grade framework for orchestrating sophisticated AI agent systems. From simple automations to complex real-world applications, CrewAI provides precise control and deep customization. By fostering collaborative intelligence through flexible, production-ready architecture, CrewAI empowers agents to work together seamlessly, tackling complex business challenges with predictable, consistent results.
+**CrewAI**: Production-grade framework for orchestrating sophisticated AI agent systems. From simple automations to complex real-world applications, CrewAI provides precise control and deep customization. By fostering collaborative intelligence through flexible, production-ready architecture, CrewAI empowers agents to work together seamlessly, tackling complex business challenges with predictable, consistent results.
+
+**CrewAI Enterprise**
+Want to plan, build (+ no code), deploy, monitor and interare your agents: [CrewAI Enterprise](https://www.crewai.com/enterprise). Designed for complex, real-world applications, our enterprise solution offers:
+
+- **Seamless Integrations**
+- **Scalable & Secure Deployment**
+- **Actionable Insights**
+- **24/7 Support**
 
 <h3>
 
@@ -190,7 +198,7 @@ research_task:
   description: >
     Conduct a thorough research about {topic}
     Make sure you find any interesting and relevant information given
-    the current year is 2024.
+    the current year is 2025.
   expected_output: >
     A list with 10 bullet points of the most relevant information about {topic}
   agent: researcher
@@ -392,7 +400,7 @@ class AdvancedAnalysisFlow(Flow[MarketState]):
             goal="Gather and validate supporting market data",
             backstory="You excel at finding and correlating multiple data sources"
         )
-        
+
         analysis_task = Task(
             description="Analyze {sector} sector data for the past {timeframe}",
             expected_output="Detailed market analysis with confidence score",
@@ -403,7 +411,7 @@ class AdvancedAnalysisFlow(Flow[MarketState]):
             expected_output="Corroborating evidence and potential contradictions",
             agent=researcher
         )
-        
+
         # Demonstrate crew autonomy
         analysis_crew = Crew(
             agents=[analyst, researcher],

docs/concepts/agents.mdx

@@ -43,7 +43,7 @@ Think of an agent as a specialized team member with specific skills, expertise,
 | **Max Retry Limit** _(optional)_        | `max_retry_limit`        | `int`                         | Maximum number of retries when an error occurs. Default is 2.                                                         |
 | **Respect Context Window** _(optional)_ | `respect_context_window` | `bool`                        | Keep messages under context window size by summarizing. Default is True.                                              |
 | **Code Execution Mode** _(optional)_    | `code_execution_mode`    | `Literal["safe", "unsafe"]`   | Mode for code execution: 'safe' (using Docker) or 'unsafe' (direct). Default is 'safe'.                               |
-| **Embedder Config** _(optional)_        | `embedder_config`        | `Optional[Dict[str, Any]]`    | Configuration for the embedder used by the agent.                                                                     |
+| **Embedder** _(optional)_               | `embedder`               | `Optional[Dict[str, Any]]`    | Configuration for the embedder used by the agent.                                                                     |
 | **Knowledge Sources** _(optional)_      | `knowledge_sources`      | `Optional[List[BaseKnowledgeSource]]` | Knowledge sources available to the agent.                                                                     |
 | **Use System Prompt** _(optional)_      | `use_system_prompt`      | `Optional[bool]`              | Whether to use system prompt (for o1 model support). Default is True.                                                 |
 
@@ -152,7 +152,7 @@ agent = Agent(
     use_system_prompt=True,  # Default: True
     tools=[SerperDevTool()],  # Optional: List of tools
     knowledge_sources=None,  # Optional: List of knowledge sources
-    embedder_config=None,  # Optional: Custom embedder configuration
+    embedder=None,  # Optional: Custom embedder configuration
     system_template=None,  # Optional: Custom system prompt template
     prompt_template=None,  # Optional: Custom prompt template
     response_template=None,  # Optional: Custom response template

docs/concepts/cli.mdx

@@ -12,7 +12,7 @@ The CrewAI CLI provides a set of commands to interact with CrewAI, allowing you
 
 To use the CrewAI CLI, make sure you have CrewAI installed:
 
-```shell
+```shell Terminal
 pip install crewai
 ```
 
@@ -20,7 +20,7 @@ pip install crewai
 
 The basic structure of a CrewAI CLI command is:
 
-```shell
+```shell Terminal
 crewai [COMMAND] [OPTIONS] [ARGUMENTS]
 ```
 
@@ -30,7 +30,7 @@ crewai [COMMAND] [OPTIONS] [ARGUMENTS]
 
 Create a new crew or flow.
 
-```shell
+```shell Terminal
 crewai create [OPTIONS] TYPE NAME
 ```
 
@@ -38,7 +38,7 @@ crewai create [OPTIONS] TYPE NAME
 - `NAME`: Name of the crew or flow
 
 Example:
-```shell
+```shell Terminal
 crewai create crew my_new_crew
 crewai create flow my_new_flow
 ```
@@ -47,14 +47,14 @@ crewai create flow my_new_flow
 
 Show the installed version of CrewAI.
 
-```shell
+```shell Terminal
 crewai version [OPTIONS]
 ```
 
 - `--tools`: (Optional) Show the installed version of CrewAI tools
 
 Example:
-```shell
+```shell Terminal
 crewai version
 crewai version --tools
 ```
@@ -63,7 +63,7 @@ crewai version --tools
 
 Train the crew for a specified number of iterations.
 
-```shell
+```shell Terminal
 crewai train [OPTIONS]
 ```
 
@@ -71,7 +71,7 @@ crewai train [OPTIONS]
 - `-f, --filename TEXT`: Path to a custom file for training (default: "trained_agents_data.pkl")
 
 Example:
-```shell
+```shell Terminal
 crewai train -n 10 -f my_training_data.pkl
 ```
 
@@ -79,14 +79,14 @@ crewai train -n 10 -f my_training_data.pkl
 
 Replay the crew execution from a specific task.
 
-```shell
+```shell Terminal
 crewai replay [OPTIONS]
 ```
 
 - `-t, --task_id TEXT`: Replay the crew from this task ID, including all subsequent tasks
 
 Example:
-```shell    
+```shell Terminal    
 crewai replay -t task_123456
 ```
 
@@ -94,7 +94,7 @@ crewai replay -t task_123456
 
 Retrieve your latest crew.kickoff() task outputs.
 
-```shell
+```shell Terminal
 crewai log-tasks-outputs
 ```
 
@@ -102,7 +102,7 @@ crewai log-tasks-outputs
 
 Reset the crew memories (long, short, entity, latest_crew_kickoff_outputs).
 
-```shell
+```shell Terminal
 crewai reset-memories [OPTIONS]
 ```
 
@@ -113,7 +113,7 @@ crewai reset-memories [OPTIONS]
 - `-a, --all`: Reset ALL memories
 
 Example:
-```shell
+```shell Terminal
 crewai reset-memories --long --short
 crewai reset-memories --all
 ```
@@ -122,7 +122,7 @@ crewai reset-memories --all
 
 Test the crew and evaluate the results.
 
-```shell
+```shell Terminal
 crewai test [OPTIONS]
 ```
 
@@ -130,7 +130,7 @@ crewai test [OPTIONS]
 - `-m, --model TEXT`: LLM Model to run the tests on the Crew (default: "gpt-4o-mini")
 
 Example:
-```shell
+```shell Terminal    
 crewai test -n 5 -m gpt-3.5-turbo
 ```
 
@@ -138,7 +138,7 @@ crewai test -n 5 -m gpt-3.5-turbo
 
 Run the crew.
 
-```shell
+```shell Terminal
 crewai run
 ```
 <Note>
@@ -147,7 +147,36 @@ Some commands may require additional configuration or setup within your project
 </Note>
 
 
-### 9. API Keys
+### 9. Chat
+
+Starting in version `0.98.0`, when you run the `crewai chat` command, you start an interactive session with your crew. The AI assistant will guide you by asking for necessary inputs to execute the crew. Once all inputs are provided, the crew will execute its tasks.
+
+After receiving the results, you can continue interacting with the assistant for further instructions or questions.
+
+```shell Terminal
+crewai chat
+```
+<Note>
+Ensure you execute these commands from your CrewAI project's root directory.
+</Note>
+<Note>
+IMPORTANT: Set the `chat_llm` property in your `crew.py` file to enable this command.
+
+```python
+@crew
+def crew(self) -> Crew:
+    return Crew(
+        agents=self.agents,
+        tasks=self.tasks,
+        process=Process.sequential,
+        verbose=True,
+        chat_llm="gpt-4o",  # LLM for chat orchestration
+    )
+```
+</Note>
+
+
+### 10. API Keys
 
 When running ```crewai create crew``` command, the CLI will first show you the top 5 most common LLM providers and ask you to select one.
 

docs/concepts/crews.mdx

@@ -23,14 +23,14 @@ A crew in crewAI represents a collaborative group of agents working together to
 | **Language** _(optional)_             | `language`             | Language used for the crew, defaults to English.                                                                                                                                                                                                          |
 | **Language File** _(optional)_        | `language_file`        | Path to the language file to be used for the crew.                                                                                                                                                                                                        |
 | **Memory** _(optional)_               | `memory`               | Utilized for storing execution memories (short-term, long-term, entity memory).                                                                                                                                                                           |
-| **Memory Config** _(optional)_        | `memory_config`        | Configuration for the memory provider to be used by the crew.                                                                                                                                                                                          |
-| **Cache** _(optional)_                | `cache`                | Specifies whether to use a cache for storing the results of tools' execution. Defaults to `True`.                                                                                                                                                          |
-| **Embedder** _(optional)_             | `embedder`             | Configuration for the embedder to be used by the crew. Mostly used by memory for now. Default is `{"provider": "openai"}`.                                                                                                                                                                     |
-| **Full Output** _(optional)_          | `full_output`          | Whether the crew should return the full output with all tasks outputs or just the final output. Defaults to `False`.                                                                                                                                       |
+| **Memory Config** _(optional)_        | `memory_config`        | Configuration for the memory provider to be used by the crew.                                                                                                                                                                                             |
+| **Cache** _(optional)_                | `cache`                | Specifies whether to use a cache for storing the results of tools' execution. Defaults to `True`.                                                                                                                                                         |
+| **Embedder** _(optional)_             | `embedder`             | Configuration for the embedder to be used by the crew. Mostly used by memory for now. Default is `{"provider": "openai"}`.                                                                                                                                |
+| **Full Output** _(optional)_          | `full_output`          | Whether the crew should return the full output with all tasks outputs or just the final output. Defaults to `False`.                                                                                                                                      |
 | **Step Callback** _(optional)_        | `step_callback`        | A function that is called after each step of every agent. This can be used to log the agent's actions or to perform other operations; it won't override the agent-specific `step_callback`.                                                               |
 | **Task Callback** _(optional)_        | `task_callback`        | A function that is called after the completion of each task. Useful for monitoring or additional operations post-task execution.                                                                                                                          |
 | **Share Crew** _(optional)_           | `share_crew`           | Whether you want to share the complete crew information and execution with the crewAI team to make the library better, and allow us to train models.                                                                                                      |
-| **Output Log File** _(optional)_      | `output_log_file`      | Whether you want to have a file with the complete crew output and execution. You can set it using True and it will default to the folder you are currently in and it will be called logs.txt or passing a string with the full path and name of the file. |
+| **Output Log File** _(optional)_      | `output_log_file`      | Set to True to save logs as logs.txt in the current directory or provide a file path. Logs will be in JSON format if the filename ends in .json, otherwise .txt. Defautls to `None`.                                                                      |
 | **Manager Agent** _(optional)_        | `manager_agent`        | `manager` sets a custom agent that will be used as a manager.                                                                                                                                                                                             |
 | **Prompt File** _(optional)_          | `prompt_file`          | Path to the prompt JSON file to be used for the crew.                                                                                                                                                                                                     |
 | **Planning** *(optional)*             | `planning`             | Adds planning ability to the Crew. When activated before each Crew iteration, all Crew data is sent to an AgentPlanner that will plan the tasks and this plan will be added to each task description.                                                     |
@@ -240,6 +240,23 @@ print(f"Tasks Output: {crew_output.tasks_output}")
 print(f"Token Usage: {crew_output.token_usage}")
 ```
 
+## Accessing Crew Logs
+
+You can see real time log of the crew execution, by setting `output_log_file` as a `True(Boolean)` or a `file_name(str)`. Supports logging of events as both `file_name.txt` and `file_name.json`.
+In case of `True(Boolean)` will save as `logs.txt`.
+
+In case of `output_log_file` is set as `False(Booelan)` or `None`, the logs will not be populated.
+
+```python Code
+# Save crew logs
+crew = Crew(output_log_file = True)  # Logs will be saved as logs.txt
+crew = Crew(output_log_file = file_name)  # Logs will be saved as file_name.txt
+crew = Crew(output_log_file = file_name.txt)  # Logs will be saved as file_name.txt
+crew = Crew(output_log_file = file_name.json)  # Logs will be saved as file_name.json
+```
+
+
+
 ## Memory Utilization
 
 Crews can utilize memory (short-term, long-term, and entity memory) to enhance their execution and learning over time. This feature allows crews to store and recall execution memories, aiding in decision-making and task execution strategies.
@@ -279,9 +296,9 @@ print(result)
 Once your crew is assembled, initiate the workflow with the appropriate kickoff method. CrewAI provides several methods for better control over the kickoff process: `kickoff()`, `kickoff_for_each()`, `kickoff_async()`, and `kickoff_for_each_async()`.
 
 - `kickoff()`: Starts the execution process according to the defined process flow.
-- `kickoff_for_each()`: Executes tasks for each agent individually.
+- `kickoff_for_each()`: Executes tasks sequentially for each provided input event or item in the collection.
 - `kickoff_async()`: Initiates the workflow asynchronously.
-- `kickoff_for_each_async()`: Executes tasks for each agent individually in an asynchronous manner.
+- `kickoff_for_each_async()`: Executes tasks concurrently for each provided input event or item, leveraging asynchronous processing.
 
 ```python Code
 # Start the crew's task execution

docs/concepts/flows.mdx

@@ -232,18 +232,18 @@ class UnstructuredExampleFlow(Flow):
     def first_method(self):
         # The state automatically includes an 'id' field
         print(f"State ID: {self.state['id']}")
-        self.state.message = "Hello from structured flow"
-        self.state.counter = 0
+        self.state['counter'] = 0
+        self.state['message'] = "Hello from structured flow"
 
     @listen(first_method)
     def second_method(self):
-        self.state.counter += 1
-        self.state.message += " - updated"
+        self.state['counter'] += 1
+        self.state['message'] += " - updated"
 
     @listen(second_method)
     def third_method(self):
-        self.state.counter += 1
-        self.state.message += " - updated again"
+        self.state['counter'] += 1
+        self.state['message'] += " - updated again"
 
         print(f"State after third_method: {self.state}")
 
@@ -323,6 +323,91 @@ flow.kickoff()
 
 By providing both unstructured and structured state management options, CrewAI Flows empowers developers to build AI workflows that are both flexible and robust, catering to a wide range of application requirements.
 
+## Flow Persistence
+
+The @persist decorator enables automatic state persistence in CrewAI Flows, allowing you to maintain flow state across restarts or different workflow executions. This decorator can be applied at either the class level or method level, providing flexibility in how you manage state persistence.
+
+### Class-Level Persistence
+
+When applied at the class level, the @persist decorator automatically persists all flow method states:
+
+```python
+@persist  # Using SQLiteFlowPersistence by default
+class MyFlow(Flow[MyState]):
+    @start()
+    def initialize_flow(self):
+        # This method will automatically have its state persisted
+        self.state.counter = 1
+        print("Initialized flow. State ID:", self.state.id)
+
+    @listen(initialize_flow)
+    def next_step(self):
+        # The state (including self.state.id) is automatically reloaded
+        self.state.counter += 1
+        print("Flow state is persisted. Counter:", self.state.counter)
+```
+
+### Method-Level Persistence
+
+For more granular control, you can apply @persist to specific methods:
+
+```python
+class AnotherFlow(Flow[dict]):
+    @persist  # Persists only this method's state
+    @start()
+    def begin(self):
+        if "runs" not in self.state:
+            self.state["runs"] = 0
+        self.state["runs"] += 1
+        print("Method-level persisted runs:", self.state["runs"])
+```
+
+### How It Works
+
+1. **Unique State Identification**
+   - Each flow state automatically receives a unique UUID
+   - The ID is preserved across state updates and method calls
+   - Supports both structured (Pydantic BaseModel) and unstructured (dictionary) states
+
+2. **Default SQLite Backend**
+   - SQLiteFlowPersistence is the default storage backend
+   - States are automatically saved to a local SQLite database
+   - Robust error handling ensures clear messages if database operations fail
+
+3. **Error Handling**
+   - Comprehensive error messages for database operations
+   - Automatic state validation during save and load
+   - Clear feedback when persistence operations encounter issues
+
+### Important Considerations
+
+- **State Types**: Both structured (Pydantic BaseModel) and unstructured (dictionary) states are supported
+- **Automatic ID**: The `id` field is automatically added if not present
+- **State Recovery**: Failed or restarted flows can automatically reload their previous state
+- **Custom Implementation**: You can provide your own FlowPersistence implementation for specialized storage needs
+
+### Technical Advantages
+
+1. **Precise Control Through Low-Level Access**
+   - Direct access to persistence operations for advanced use cases
+   - Fine-grained control via method-level persistence decorators
+   - Built-in state inspection and debugging capabilities
+   - Full visibility into state changes and persistence operations
+
+2. **Enhanced Reliability**
+   - Automatic state recovery after system failures or restarts
+   - Transaction-based state updates for data integrity
+   - Comprehensive error handling with clear error messages
+   - Robust validation during state save and load operations
+
+3. **Extensible Architecture**
+   - Customizable persistence backend through FlowPersistence interface
+   - Support for specialized storage solutions beyond SQLite
+   - Compatible with both structured (Pydantic) and unstructured (dict) states
+   - Seamless integration with existing CrewAI flow patterns
+
+The persistence system's architecture emphasizes technical precision and customization options, allowing developers to maintain full control over state management while benefiting from built-in reliability features.
+
 ## Flow Control
 
 ### Conditional Logic: `or`

docs/concepts/knowledge.mdx

@@ -93,6 +93,12 @@ result = crew.kickoff(inputs={"question": "What city does John live in and how o
 
 Here's another example with the `CrewDoclingSource`. The CrewDoclingSource is actually quite versatile and can handle multiple file formats including TXT, PDF, DOCX, HTML, and more. 
 
+<Note>
+  You need to install `docling` for the following example to work: `uv add docling`
+</Note>
+
+
+
 ```python Code
 from crewai import LLM, Agent, Crew, Process, Task
 from crewai.knowledge.source.crew_docling_source import CrewDoclingSource
@@ -282,6 +288,7 @@ The `embedder` parameter supports various embedding model providers that include
 - `ollama`: Local embeddings with Ollama
 - `vertexai`: Google Cloud VertexAI embeddings
 - `cohere`: Cohere's embedding models
+- `voyageai`: VoyageAI's embedding models
 - `bedrock`: AWS Bedrock embeddings
 - `huggingface`: Hugging Face models
 - `watson`: IBM Watson embeddings
@@ -317,6 +324,13 @@ agent = Agent(
     verbose=True,
     allow_delegation=False,
     llm=gemini_llm,
+    embedder={
+        "provider": "google",
+        "config": {
+            "model": "models/text-embedding-004",
+            "api_key": GEMINI_API_KEY,
+        }
+    }
 )
 
 task = Task(

docs/concepts/llms.mdx

@@ -38,6 +38,7 @@ Here's a detailed breakdown of supported models and their capabilities, you can
     | GPT-4 | 8,192 tokens | High-accuracy tasks, complex reasoning |
     | GPT-4 Turbo | 128,000 tokens | Long-form content, document analysis |
     | GPT-4o & GPT-4o-mini | 128,000 tokens | Cost-effective large context processing |
+    | o3-mini | 200,000 tokens | Fast reasoning, complex reasoning |
 
     <Note>
       1 token ≈ 4 characters in English. For example, 8,192 tokens ≈ 32,768 characters or about 6,000 words.
@@ -162,7 +163,8 @@ Here's a detailed breakdown of supported models and their capabilities, you can
   <Tab title="Others">
     | Provider | Context Window | Key Features |
     |----------|---------------|--------------|
-    | Deepseek Chat | 128,000 tokens | Specialized in technical discussions |
+    | Deepseek Chat | 64,000 tokens | Specialized in technical discussions |
+    | Deepseek R1 | 64,000 tokens | Affordable reasoning model |
     | Claude 3 | Up to 200K tokens | Strong reasoning, code understanding |
     | Gemma Series | 8,192 tokens | Efficient, smaller-scale tasks |
 
@@ -243,6 +245,9 @@ There are three ways to configure LLMs in CrewAI. Choose the method that best fi
         # llm: bedrock/amazon.titan-text-express-v1
         # llm: bedrock/meta.llama2-70b-chat-v1
 
+        # Amazon SageMaker Models - Enterprise-grade
+        # llm: sagemaker/<my-endpoint>
+
         # Mistral Models - Open source alternative
         # llm: mistral/mistral-large-latest
         # llm: mistral/mistral-medium-latest
@@ -293,6 +298,10 @@ There are three ways to configure LLMs in CrewAI. Choose the method that best fi
         # llm: sambanova/Meta-Llama-3.1-8B-Instruct
         # llm: sambanova/BioMistral-7B
         # llm: sambanova/Falcon-180B
+
+        # Open Router Models - Affordable reasoning
+        # llm: openrouter/deepseek/deepseek-r1
+        # llm: openrouter/deepseek/deepseek-chat
     ```
 
     <Info>
@@ -454,19 +463,36 @@ Learn how to get the most out of your LLM configuration:
 
   <Accordion title="Google">
     ```python Code
-    # Option 1. Gemini accessed with an API key.
+    # Option 1: Gemini accessed with an API key.
     # https://ai.google.dev/gemini-api/docs/api-key
     GEMINI_API_KEY=<your-api-key>
 
-    # Option 2. Vertex AI IAM credentials for Gemini, Anthropic, and anything in the Model Garden.
+    # Option 2: Vertex AI IAM credentials for Gemini, Anthropic, and Model Garden.
     # https://cloud.google.com/vertex-ai/generative-ai/docs/overview
     ```
 
+    Get credentials:
+    ```python Code
+    import json
+
+    file_path = 'path/to/vertex_ai_service_account.json'
+
+    # Load the JSON file
+    with open(file_path, 'r') as file:
+        vertex_credentials = json.load(file)
+
+    # Convert the credentials to a JSON string
+    vertex_credentials_json = json.dumps(vertex_credentials)
+    ```
+
     Example usage:
     ```python Code
+    from crewai import LLM
+
     llm = LLM(
         model="gemini/gemini-1.5-pro-latest",
-        temperature=0.7
+        temperature=0.7,
+        vertex_credentials=vertex_credentials_json
     )
     ```
   </Accordion>
@@ -506,6 +532,21 @@ Learn how to get the most out of your LLM configuration:
     )
     ```
   </Accordion>
+  
+  <Accordion title="Amazon SageMaker">
+    ```python Code
+    AWS_ACCESS_KEY_ID=<your-access-key>
+    AWS_SECRET_ACCESS_KEY=<your-secret-key>
+    AWS_DEFAULT_REGION=<your-region>
+    ```
+
+    Example usage:
+    ```python Code
+    llm = LLM(
+        model="sagemaker/<my-endpoint>"
+    )
+    ```
+  </Accordion>
 
   <Accordion title="Mistral">
     ```python Code
@@ -662,8 +703,53 @@ Learn how to get the most out of your LLM configuration:
       - Support for long context windows
     </Info>
   </Accordion>
+
+  <Accordion title="Open Router">
+    ```python Code
+    OPENROUTER_API_KEY=<your-api-key>
+    ```
+    
+    Example usage:
+    ```python Code
+    llm = LLM(
+        model="openrouter/deepseek/deepseek-r1",
+        base_url="https://openrouter.ai/api/v1",
+        api_key=OPENROUTER_API_KEY
+    )
+    ```
+
+    <Info>
+      Open Router models:
+      - openrouter/deepseek/deepseek-r1
+      - openrouter/deepseek/deepseek-chat
+    </Info>
+  </Accordion>
 </AccordionGroup>
 
+## Structured LLM Calls
+
+CrewAI supports structured responses from LLM calls by allowing you to define a `response_format` using a Pydantic model. This enables the framework to automatically parse and validate the output, making it easier to integrate the response into your application without manual post-processing.
+
+For example, you can define a Pydantic model to represent the expected response structure and pass it as the `response_format` when instantiating the LLM. The model will then be used to convert the LLM output into a structured Python object.
+
+```python Code
+from crewai import LLM
+
+class Dog(BaseModel):
+    name: str
+    age: int
+    breed: str
+
+
+llm = LLM(model="gpt-4o", response_format=Dog)
+
+response = llm.call(
+    "Analyze the following messages and return the name, age, and breed. "
+    "Meet Kona! She is 3 years old and is a black german shepherd."
+)
+print(response)
+```
+
 ## Common Issues and Solutions
 
 <Tabs>

docs/concepts/memory.mdx

@@ -58,41 +58,107 @@ my_crew = Crew(
 ### Example: Use Custom Memory Instances e.g FAISS as the VectorDB
 
 ```python Code
-from crewai import Crew, Agent, Task, Process
+from crewai import Crew, Process
+from crewai.memory import LongTermMemory, ShortTermMemory, EntityMemory
+from crewai.memory.storage import LTMSQLiteStorage, RAGStorage
+from typing import List, Optional
 
 # Assemble your crew with memory capabilities
-my_crew = Crew(
-    agents=[...],
-    tasks=[...],
-    process="Process.sequential",
-    memory=True,
-    long_term_memory=EnhanceLongTermMemory(
+my_crew: Crew = Crew(
+    agents = [...],
+    tasks = [...],
+    process = Process.sequential,
+    memory = True,
+    # Long-term memory for persistent storage across sessions
+    long_term_memory = LongTermMemory(
         storage=LTMSQLiteStorage(
-            db_path="/my_data_dir/my_crew1/long_term_memory_storage.db"
+            db_path="/my_crew1/long_term_memory_storage.db"
         )
     ),
-    short_term_memory=EnhanceShortTermMemory(
-        storage=CustomRAGStorage(
-            crew_name="my_crew",
-            storage_type="short_term",
-            data_dir="//my_data_dir",
-            model=embedder["model"],
-            dimension=embedder["dimension"],
+    # Short-term memory for current context using RAG
+    short_term_memory = ShortTermMemory(
+        storage = RAGStorage(
+                embedder_config={
+                    "provider": "openai",
+                    "config": {
+                        "model": 'text-embedding-3-small'
+                    }
+                },
+                type="short_term",
+                path="/my_crew1/"
+            )
         ),
     ),
-    entity_memory=EnhanceEntityMemory(
-        storage=CustomRAGStorage(
-            crew_name="my_crew",
-            storage_type="entities",
-            data_dir="//my_data_dir",
-            model=embedder["model"],
-            dimension=embedder["dimension"],
-        ),
+    # Entity memory for tracking key information about entities
+    entity_memory = EntityMemory(
+        storage=RAGStorage(
+            embedder_config={
+                "provider": "openai",
+                "config": {
+                    "model": 'text-embedding-3-small'
+                }
+            },
+            type="short_term",
+            path="/my_crew1/"
+        )
     ),
     verbose=True,
 )
 ```
 
+## Security Considerations
+
+When configuring memory storage:
+- Use environment variables for storage paths (e.g., `CREWAI_STORAGE_DIR`)
+- Never hardcode sensitive information like database credentials
+- Consider access permissions for storage directories
+- Use relative paths when possible to maintain portability
+
+Example using environment variables:
+```python
+import os
+from crewai import Crew
+from crewai.memory import LongTermMemory
+from crewai.memory.storage import LTMSQLiteStorage
+
+# Configure storage path using environment variable
+storage_path = os.getenv("CREWAI_STORAGE_DIR", "./storage")
+crew = Crew(
+    memory=True,
+    long_term_memory=LongTermMemory(
+        storage=LTMSQLiteStorage(
+            db_path="{storage_path}/memory.db".format(storage_path=storage_path)
+        )
+    )
+)
+```
+
+## Configuration Examples
+
+### Basic Memory Configuration
+```python
+from crewai import Crew
+from crewai.memory import LongTermMemory
+
+# Simple memory configuration
+crew = Crew(memory=True)  # Uses default storage locations
+```
+
+### Custom Storage Configuration
+```python
+from crewai import Crew
+from crewai.memory import LongTermMemory
+from crewai.memory.storage import LTMSQLiteStorage
+
+# Configure custom storage paths
+crew = Crew(
+    memory=True,
+    long_term_memory=LongTermMemory(
+        storage=LTMSQLiteStorage(db_path="./memory.db")
+    )
+)
+```
+
 ## Integrating Mem0 for Enhanced User Memory
 
 [Mem0](https://mem0.ai/) is a self-improving memory layer for LLM applications, enabling personalized AI experiences. 
@@ -185,7 +251,12 @@ my_crew = Crew(
     process=Process.sequential,
     memory=True,
     verbose=True,
-    embedder=OpenAIEmbeddingFunction(api_key=os.getenv("OPENAI_API_KEY"), model_name="text-embedding-3-small"),
+    embedder={
+        "provider": "openai",
+        "config": {
+            "model": 'text-embedding-3-small'
+        }
+    }
 )
 ```
 
@@ -224,7 +295,7 @@ my_crew = Crew(
         "provider": "google",
         "config": {
             "api_key": "<YOUR_API_KEY>",
-            "model_name": "<model_name>"
+            "model": "<model_name>"
         }
     }
 )
@@ -242,13 +313,15 @@ my_crew = Crew(
     process=Process.sequential,
     memory=True,
     verbose=True,
-    embedder=OpenAIEmbeddingFunction(
-        api_key="YOUR_API_KEY",
-        api_base="YOUR_API_BASE_PATH",
-        api_type="azure",
-        api_version="YOUR_API_VERSION",
-        model_name="text-embedding-3-small"
-    )
+    embedder={
+        "provider": "openai",
+        "config": {
+            "api_key": "YOUR_API_KEY",
+            "api_base": "YOUR_API_BASE_PATH",
+            "api_version": "YOUR_API_VERSION",
+            "model_name": 'text-embedding-3-small'
+        }
+    }
 )
 ```
 
@@ -264,12 +337,15 @@ my_crew = Crew(
     process=Process.sequential,
     memory=True,
     verbose=True,
-    embedder=GoogleVertexEmbeddingFunction(
-        project_id="YOUR_PROJECT_ID",
-        region="YOUR_REGION",
-        api_key="YOUR_API_KEY",
-        model_name="textembedding-gecko"
-    )
+    embedder={
+        "provider": "vertexai",
+        "config": {
+            "project_id"="YOUR_PROJECT_ID",
+            "region"="YOUR_REGION",
+            "api_key"="YOUR_API_KEY",
+            "model_name"="textembedding-gecko"
+        }
+    }
 )
 ```
 
@@ -288,7 +364,27 @@ my_crew = Crew(
         "provider": "cohere",
         "config": {
             "api_key": "YOUR_API_KEY",
-            "model_name": "<model_name>"
+            "model": "<model_name>"
+        }
+    }
+)
+```
+### Using VoyageAI embeddings
+
+```python Code
+from crewai import Crew, Agent, Task, Process
+
+my_crew = Crew(
+    agents=[...],
+    tasks=[...],
+    process=Process.sequential,
+    memory=True,
+    verbose=True,
+    embedder={
+        "provider": "voyageai",
+        "config": {
+            "api_key": "YOUR_API_KEY",
+            "model": "<model_name>"
         }
     }
 )
@@ -338,6 +434,33 @@ my_crew = Crew(
 )
 ```
 
+### Adding Custom Embedding Function
+
+```python Code
+from crewai import Crew, Agent, Task, Process
+from chromadb import Documents, EmbeddingFunction, Embeddings
+
+# Create a custom embedding function
+class CustomEmbedder(EmbeddingFunction):
+    def __call__(self, input: Documents) -> Embeddings:
+        # generate embeddings
+        return [1, 2, 3] # this is a dummy embedding
+
+my_crew = Crew(
+    agents=[...],
+    tasks=[...],
+    process=Process.sequential,
+    memory=True,
+    verbose=True,
+    embedder={
+        "provider": "custom",
+        "config": {
+            "embedder": CustomEmbedder()
+        }
+    }
+)
+```
+
 ### Resetting Memory
 
 ```shell

docs/concepts/planning.mdx

@@ -81,8 +81,8 @@ my_crew.kickoff()
 
 3. **Collect Data:**
 
-   - Search for the latest papers, articles, and reports published in 2023 and early 2024.
-   - Use keywords like "Large Language Models 2024", "AI LLM advancements", "AI ethics 2024", etc.
+   - Search for the latest papers, articles, and reports published in 2024 and early 2025.
+   - Use keywords like "Large Language Models 2025", "AI LLM advancements", "AI ethics 2025", etc.
 
 4. **Analyze Findings:**
 

docs/concepts/tasks.mdx

@@ -33,11 +33,12 @@ crew = Crew(
 | :------------------------------- | :---------------- | :---------------------------- | :------------------------------------------------------------------------------------------------------------------- |
 | **Description**                  | `description`     | `str`                         | A clear, concise statement of what the task entails.                                                                 |
 | **Expected Output**              | `expected_output` | `str`                         | A detailed description of what the task's completion looks like.                                                     |
-| **Name** _(optional)_           | `name`           | `Optional[str]`               | A name identifier for the task.                                                                                      |
-| **Agent** _(optional)_          | `agent`           | `Optional[BaseAgent]`         | The agent responsible for executing the task.                                                                        |
-| **Tools** _(optional)_          | `tools`           | `List[BaseTool]`             | The tools/resources the agent is limited to use for this task.                                                       |
+| **Name** _(optional)_            | `name`            | `Optional[str]`               | A name identifier for the task.                                                                                      |
+| **Agent** _(optional)_           | `agent`           | `Optional[BaseAgent]`         | The agent responsible for executing the task.                                                                        |
+| **Tools** _(optional)_           | `tools`           | `List[BaseTool]`              | The tools/resources the agent is limited to use for this task.                                                       |
 | **Context** _(optional)_         | `context`         | `Optional[List["Task"]]`      | Other tasks whose outputs will be used as context for this task.                                                     |
 | **Async Execution** _(optional)_ | `async_execution` | `Optional[bool]`              | Whether the task should be executed asynchronously. Defaults to False.                                               |
+| **Human Input** _(optional)_     | `human_input`     | `Optional[bool]`              | Whether the task should have a human review the final answer of the agent. Defaults to False.                        |
 | **Config** _(optional)_          | `config`          | `Optional[Dict[str, Any]]`    | Task-specific configuration parameters.                                                                              |
 | **Output File** _(optional)_     | `output_file`     | `Optional[str]`               | File path for storing the task output.                                                                               |
 | **Output JSON** _(optional)_     | `output_json`     | `Optional[Type[BaseModel]]`   | A Pydantic model to structure the JSON output.                                                                       |
@@ -68,7 +69,7 @@ research_task:
   description: >
     Conduct a thorough research about {topic}
     Make sure you find any interesting and relevant information given
-    the current year is 2024.
+    the current year is 2025.
   expected_output: >
     A list with 10 bullet points of the most relevant information about {topic}
   agent: researcher
@@ -154,7 +155,7 @@ research_task = Task(
     description="""
         Conduct a thorough research about AI Agents.
         Make sure you find any interesting and relevant information given
-        the current year is 2024.
+        the current year is 2025.
     """,
     expected_output="""
         A list with 10 bullet points of the most relevant information about AI Agents

docs/how-to/human-input-on-execution.mdx

@@ -60,12 +60,12 @@ writer = Agent(
 # Create tasks for your agents
 task1 = Task(
     description=(
-        "Conduct a comprehensive analysis of the latest advancements in AI in 2024. "
+        "Conduct a comprehensive analysis of the latest advancements in AI in 2025. "
         "Identify key trends, breakthrough technologies, and potential industry impacts. "
         "Compile your findings in a detailed report. "
         "Make sure to check with a human if the draft is good before finalizing your answer."
     ),
-    expected_output='A comprehensive full report on the latest AI advancements in 2024, leave nothing out',
+    expected_output='A comprehensive full report on the latest AI advancements in 2025, leave nothing out',
     agent=researcher,
     human_input=True
 )
@@ -76,7 +76,7 @@ task2 = Task(
         "Your post should be informative yet accessible, catering to a tech-savvy audience. "
         "Aim for a narrative that captures the essence of these breakthroughs and their implications for the future."
     ),
-    expected_output='A compelling 3 paragraphs blog post formatted as markdown about the latest AI advancements in 2024',
+    expected_output='A compelling 3 paragraphs blog post formatted as markdown about the latest AI advancements in 2025',
     agent=writer,
     human_input=True
 )

docs/how-to/llm-connections.mdx

@@ -23,6 +23,7 @@ LiteLLM supports a wide range of providers, including but not limited to:
 - Azure OpenAI
 - AWS (Bedrock, SageMaker)
 - Cohere
+- VoyageAI
 - Hugging Face
 - Ollama
 - Mistral AI

docs/how-to/mlflow-observability.mdx

@@ -0,0 +1,206 @@
+---
+title: Agent Monitoring with MLflow
+description: Quickly start monitoring your Agents with MLflow.
+icon: bars-staggered
+---
+
+# MLflow Overview
+
+[MLflow](https://mlflow.org/) is an open-source platform to assist machine learning practitioners and teams in handling the complexities of the machine learning process.
+
+It provides a tracing feature that enhances LLM observability in your Generative AI applications by capturing detailed information about the execution of your application’s services. 
+Tracing provides a way to record the inputs, outputs, and metadata associated with each intermediate step of a request, enabling you to easily pinpoint the source of bugs and unexpected behaviors.
+
+![Overview of MLflow crewAI tracing usage](/images/mlflow-tracing.gif)
+
+### Features
+
+- **Tracing Dashboard**: Monitor activities of your crewAI agents with detailed dashboards that include inputs, outputs and metadata of spans.
+- **Automated Tracing**: A fully automated integration with crewAI, which can be enabled by running `mlflow.crewai.autolog()`. 
+- **Manual Trace Instrumentation with minor efforts**: Customize trace instrumentation through MLflow's high-level fluent APIs such as decorators, function wrappers and context managers.
+- **OpenTelemetry Compatibility**: MLflow Tracing supports exporting traces to an OpenTelemetry Collector, which can then be used to export traces to various backends such as Jaeger, Zipkin, and AWS X-Ray.
+- **Package and Deploy Agents**: Package and deploy your crewAI agents to an inference server with a variety of deployment targets.
+- **Securely Host LLMs**: Host multiple LLM from various providers in one unified endpoint through MFflow gateway.
+- **Evaluation**: Evaluate your crewAI agents with a wide range of metrics using a convenient API `mlflow.evaluate()`.
+
+## Setup Instructions
+
+<Steps>
+    <Step title="Install MLflow package">
+      ```shell
+      # The crewAI integration is available in mlflow>=2.19.0
+      pip install mlflow
+      ```
+    </Step>
+    <Step title="Start MFflow tracking server">
+      ```shell
+      # This process is optional, but it is recommended to use MLflow tracking server for better visualization and broader features.
+      mlflow server
+      ```
+    </Step>
+    <Step title="Initialize MLflow in Your Application">
+      Add the following two lines to your application code:
+
+      ```python
+      import mlflow
+
+      mlflow.crewai.autolog()
+
+      # Optional: Set a tracking URI and an experiment name if you have a tracking server
+      mlflow.set_tracking_uri("http://localhost:5000")
+      mlflow.set_experiment("CrewAI")
+      ```
+      
+      Example Usage for tracing CrewAI Agents:
+
+      ```python
+      from crewai import Agent, Crew, Task
+      from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
+      from crewai_tools import SerperDevTool, WebsiteSearchTool
+
+      from textwrap import dedent
+
+      content = "Users name is John. He is 30 years old and lives in San Francisco."
+      string_source = StringKnowledgeSource(
+          content=content, metadata={"preference": "personal"}
+      )
+
+      search_tool = WebsiteSearchTool()
+
+
+      class TripAgents:
+          def city_selection_agent(self):
+              return Agent(
+                  role="City Selection Expert",
+                  goal="Select the best city based on weather, season, and prices",
+                  backstory="An expert in analyzing travel data to pick ideal destinations",
+                  tools=[
+                      search_tool,
+                  ],
+                  verbose=True,
+              )
+
+          def local_expert(self):
+              return Agent(
+                  role="Local Expert at this city",
+                  goal="Provide the BEST insights about the selected city",
+                  backstory="""A knowledgeable local guide with extensive information
+              about the city, it's attractions and customs""",
+                  tools=[search_tool],
+                  verbose=True,
+              )
+
+
+      class TripTasks:
+          def identify_task(self, agent, origin, cities, interests, range):
+              return Task(
+                  description=dedent(
+                      f"""
+                      Analyze and select the best city for the trip based
+                      on specific criteria such as weather patterns, seasonal
+                      events, and travel costs. This task involves comparing
+                      multiple cities, considering factors like current weather
+                      conditions, upcoming cultural or seasonal events, and
+                      overall travel expenses.
+                      Your final answer must be a detailed
+                      report on the chosen city, and everything you found out
+                      about it, including the actual flight costs, weather
+                      forecast and attractions.
+
+                      Traveling from: {origin}
+                      City Options: {cities}
+                      Trip Date: {range}
+                      Traveler Interests: {interests}
+                  """
+                  ),
+                  agent=agent,
+                  expected_output="Detailed report on the chosen city including flight costs, weather forecast, and attractions",
+              )
+
+          def gather_task(self, agent, origin, interests, range):
+              return Task(
+                  description=dedent(
+                      f"""
+                      As a local expert on this city you must compile an
+                      in-depth guide for someone traveling there and wanting
+                      to have THE BEST trip ever!
+                      Gather information about key attractions, local customs,
+                      special events, and daily activity recommendations.
+                      Find the best spots to go to, the kind of place only a
+                      local would know.
+                      This guide should provide a thorough overview of what
+                      the city has to offer, including hidden gems, cultural
+                      hotspots, must-visit landmarks, weather forecasts, and
+                      high level costs.
+                      The final answer must be a comprehensive city guide,
+                      rich in cultural insights and practical tips,
+                      tailored to enhance the travel experience.
+
+                      Trip Date: {range}
+                      Traveling from: {origin}
+                      Traveler Interests: {interests}
+                  """
+                  ),
+                  agent=agent,
+                  expected_output="Comprehensive city guide including hidden gems, cultural hotspots, and practical travel tips",
+              )
+
+
+      class TripCrew:
+          def __init__(self, origin, cities, date_range, interests):
+              self.cities = cities
+              self.origin = origin
+              self.interests = interests
+              self.date_range = date_range
+
+          def run(self):
+              agents = TripAgents()
+              tasks = TripTasks()
+
+              city_selector_agent = agents.city_selection_agent()
+              local_expert_agent = agents.local_expert()
+
+              identify_task = tasks.identify_task(
+                  city_selector_agent,
+                  self.origin,
+                  self.cities,
+                  self.interests,
+                  self.date_range,
+              )
+              gather_task = tasks.gather_task(
+                  local_expert_agent, self.origin, self.interests, self.date_range
+              )
+
+              crew = Crew(
+                  agents=[city_selector_agent, local_expert_agent],
+                  tasks=[identify_task, gather_task],
+                  verbose=True,
+                  memory=True,
+                  knowledge={
+                      "sources": [string_source],
+                      "metadata": {"preference": "personal"},
+                  },
+              )
+
+              result = crew.kickoff()
+              return result
+
+
+      trip_crew = TripCrew("California", "Tokyo", "Dec 12 - Dec 20", "sports")
+      result = trip_crew.run()
+
+      print(result)
+      ```
+      Refer to [MLflow Tracing Documentation](https://mlflow.org/docs/latest/llms/tracing/index.html) for more configurations and use cases.
+    </Step>
+    <Step title="Visualize Activities of Agents">
+      Now traces for your crewAI agents are captured by MLflow. 
+      Let's visit MLflow tracking server to view the traces and get insights into your Agents.
+
+      Open `127.0.0.1:5000` on your browser to visit MLflow tracking server. 
+      <Frame caption="MLflow Tracing Dashboard">
+        <img src="/images/mlflow1.png" alt="MLflow tracing example with crewai" />
+      </Frame>
+    </Step>
+</Steps> 
+

docs/how-to/multimodal-agents.mdx

@@ -45,6 +45,7 @@ image_analyst = Agent(
 # Create a task for image analysis
 task = Task(
     description="Analyze the product image at https://example.com/product.jpg and provide a detailed description",
+    expected_output="A detailed description of the product image",
     agent=image_analyst
 )
 
@@ -81,6 +82,7 @@ inspection_task = Task(
     3. Compliance with standards
     Provide a detailed report highlighting any issues found.
     """,
+    expected_output="A detailed report highlighting any issues found",
     agent=expert_analyst
 )
 

docs/installation.mdx

@@ -15,10 +15,48 @@ icon: wrench
   If you need to update Python, visit [python.org/downloads](https://python.org/downloads)
 </Note>
 
+# Setting Up Your Environment
+
+Before installing CrewAI, it's recommended to set up a virtual environment. This helps isolate your project dependencies and avoid conflicts.
+
+<Steps>
+    <Step title="Create a Virtual Environment">
+        Choose your preferred method to create a virtual environment:
+
+        **Using venv (Python's built-in tool):**
+        ```shell Terminal
+        python3 -m venv .venv
+        ```
+
+        **Using conda:**
+        ```shell Terminal
+        conda create -n crewai-env python=3.12
+        ```
+    </Step>
+
+    <Step title="Activate the Virtual Environment">
+        Activate your virtual environment based on your platform:
+
+        **On macOS/Linux (venv):**
+        ```shell Terminal
+        source .venv/bin/activate
+        ```
+
+        **On Windows (venv):**
+        ```shell Terminal
+        .venv\Scripts\activate
+        ```
+
+        **Using conda (all platforms):**
+        ```shell Terminal
+        conda activate crewai-env
+        ```
+    </Step>
+</Steps>
+
 # Installing CrewAI
 
-CrewAI is a flexible and powerful AI framework that enables you to create and manage AI agents, tools, and tasks efficiently. 
-Let's get you set up! 🚀
+Now let's get you set up! 🚀
 
 <Steps>
     <Step title="Install CrewAI">
@@ -72,9 +110,9 @@ Let's get you set up! 🚀
 
 # Creating a New Project
 
-<Info>
+<Tip>
   We recommend using the YAML Template scaffolding for a structured approach to defining agents and tasks.
-</Info>
+</Tip>
 
 <Steps>
   <Step title="Generate Project Structure">
@@ -104,7 +142,18 @@ Let's get you set up! 🚀
                 └── tasks.yaml
     ```
     </Frame>
-  </Step>   
+  </Step>
+
+  <Step title="Install Additional Tools">
+    You can install additional tools using UV:
+    ```shell Terminal
+    uv add <tool-name>
+    ```
+
+    <Tip>
+      UV is our preferred package manager as it's significantly faster than pip and provides better dependency resolution.
+    </Tip>
+  </Step>
 
   <Step title="Customize Your Project">
     Your project will contain these essential files:

docs/memory.md

@@ -0,0 +1,45 @@
+# Memory in CrewAI
+
+CrewAI provides a robust memory system that allows agents to retain and recall information from previous interactions.
+
+## Configuring Embedding Providers
+
+CrewAI supports multiple embedding providers for memory functionality:
+
+- OpenAI (default) - Requires `OPENAI_API_KEY`
+- Ollama - Requires `CREWAI_OLLAMA_URL` (defaults to "http://localhost:11434/api/embeddings")
+
+### Environment Variables
+
+Configure the embedding provider using these environment variables:
+
+- `CREWAI_EMBEDDING_PROVIDER`: Provider name (default: "openai")
+- `CREWAI_EMBEDDING_MODEL`: Model name (default: "text-embedding-3-small")
+- `CREWAI_OLLAMA_URL`: URL for Ollama API (when using Ollama provider)
+
+### Example Configuration
+
+```python
+# Using OpenAI (default)
+os.environ["OPENAI_API_KEY"] = "your-api-key"
+
+# Using Ollama
+os.environ["CREWAI_EMBEDDING_PROVIDER"] = "ollama"
+os.environ["CREWAI_EMBEDDING_MODEL"] = "llama2"  # or any other model supported by your Ollama instance
+os.environ["CREWAI_OLLAMA_URL"] = "http://localhost:11434/api/embeddings"  # optional, this is the default
+```
+
+## Memory Usage
+
+When an agent has memory enabled, it can access and store information from previous interactions:
+
+```python
+agent = Agent(
+    role="Researcher",
+    goal="Research AI topics",
+    backstory="You're an AI researcher",
+    memory=True  # Enable memory for this agent
+)
+```
+
+The memory system uses embeddings to store and retrieve relevant information, allowing agents to maintain context across multiple interactions and tasks.

docs/mint.json

@@ -101,6 +101,7 @@
         "how-to/conditional-tasks",
         "how-to/agentops-observability",
         "how-to/langtrace-observability",
+        "how-to/mlflow-observability",
         "how-to/openlit-observability",
         "how-to/portkey-observability"
       ]

docs/quickstart.mdx

@@ -58,7 +58,7 @@ Follow the steps below to get crewing! 🚣‍♂️
       description: >
         Conduct a thorough research about {topic}
         Make sure you find any interesting and relevant information given
-        the current year is 2024.
+        the current year is 2025.
       expected_output: >
         A list with 10 bullet points of the most relevant information about {topic}
       agent: researcher
@@ -195,10 +195,10 @@ Follow the steps below to get crewing! 🚣‍♂️
 
   <CodeGroup>
     ```markdown output/report.md
-    # Comprehensive Report on the Rise and Impact of AI Agents in 2024
+    # Comprehensive Report on the Rise and Impact of AI Agents in 2025
 
     ## 1. Introduction to AI Agents
-    In 2024, Artificial Intelligence (AI) agents are at the forefront of innovation across various industries. As intelligent systems that can perform tasks typically requiring human cognition, AI agents are paving the way for significant advancements in operational efficiency, decision-making, and overall productivity within sectors like Human Resources (HR) and Finance. This report aims to detail the rise of AI agents, their frameworks, applications, and potential implications on the workforce.
+    In 2025, Artificial Intelligence (AI) agents are at the forefront of innovation across various industries. As intelligent systems that can perform tasks typically requiring human cognition, AI agents are paving the way for significant advancements in operational efficiency, decision-making, and overall productivity within sectors like Human Resources (HR) and Finance. This report aims to detail the rise of AI agents, their frameworks, applications, and potential implications on the workforce.
 
     ## 2. Benefits of AI Agents
     AI agents bring numerous advantages that are transforming traditional work environments. Key benefits include:
@@ -252,7 +252,7 @@ Follow the steps below to get crewing! 🚣‍♂️
     To stay competitive and harness the full potential of AI agents, organizations must remain vigilant about latest developments in AI technology and consider continuous learning and adaptation in their strategic planning.
 
     ## 8. Conclusion
-    The emergence of AI agents is undeniably reshaping the workplace landscape in 2024. With their ability to automate tasks, enhance efficiency, and improve decision-making, AI agents are critical in driving operational success. Organizations must embrace and adapt to AI developments to thrive in an increasingly digital business environment.
+    The emergence of AI agents is undeniably reshaping the workplace landscape in 5. With their ability to automate tasks, enhance efficiency, and improve decision-making, AI agents are critical in driving operational success. Organizations must embrace and adapt to AI developments to thrive in an increasingly digital business environment.
     ```
   </CodeGroup>
   </Step>
@@ -278,7 +278,7 @@ email_summarizer:
       Summarize emails into a concise and clear summary
     backstory: >
       You will create a 5 bullet point summary of the report
-    llm: mixtal_llm
+    llm: openai/gpt-4o
 ```
 
 <Tip>

docs/tools/composiotool.mdx

@@ -1,78 +1,118 @@
 ---
 title: Composio Tool
-description: The `ComposioTool` is a wrapper around the composio set of tools and gives your agent access to a wide variety of tools from the Composio SDK.
+description: Composio provides 250+ production-ready tools for AI agents with flexible authentication management.
 icon: gear-code
 ---
 
-# `ComposioTool`
+# `ComposioToolSet`
 
 ## Description
+Composio is an integration platform that allows you to connect your AI agents to 250+ tools. Key features include:
 
-This tools is a wrapper around the composio set of tools and gives your agent access to a wide variety of tools from the Composio SDK.
+- **Enterprise-Grade Authentication**: Built-in support for OAuth, API Keys, JWT with automatic token refresh
+- **Full Observability**: Detailed tool usage logs, execution timestamps, and more
 
 ## Installation
 
-To incorporate this tool into your project, follow the installation instructions below:
+To incorporate Composio tools into your project, follow the instructions below:
 
 ```shell
-pip install composio-core
-pip install 'crewai[tools]'
+pip install composio-crewai
+pip install crewai
 ```
 
-after the installation is complete, either run `composio login` or export your composio API key as `COMPOSIO_API_KEY`.
+After the installation is complete, either run `composio login` or export your composio API key as `COMPOSIO_API_KEY`. Get your Composio API key from [here](https://app.composio.dev)
 
 ## Example
 
 The following example demonstrates how to initialize the tool and execute a github action:
 
-1. Initialize Composio tools
+1. Initialize Composio toolset
 
 ```python Code
-from composio import App
-from crewai_tools import ComposioTool
-from crewai import Agent, Task
+from composio_crewai import ComposioToolSet, App, Action
+from crewai import Agent, Task, Crew
 
-
-tools = [ComposioTool.from_action(action=Action.GITHUB_ACTIVITY_STAR_REPO_FOR_AUTHENTICATED_USER)]
+toolset = ComposioToolSet()
 ```
 
-If you don't know what action you want to use, use `from_app` and `tags` filter to get relevant actions
-
+2. Connect your GitHub account
+<CodeGroup>
+```shell CLI
+composio add github
+```
 ```python Code
-tools = ComposioTool.from_app(App.GITHUB, tags=["important"])
+request = toolset.initiate_connection(app=App.GITHUB)
+print(f"Open this URL to authenticate: {request.redirectUrl}")
 ```
+</CodeGroup>
 
-or use `use_case` to search relevant actions
+3. Get Tools
 
+- Retrieving all the tools from an app (not recommended for production):
 ```python Code
-tools = ComposioTool.from_app(App.GITHUB, use_case="Star a github repository")
+tools = toolset.get_tools(apps=[App.GITHUB])
 ```
 
-2. Define agent
+- Filtering tools based on tags:
+```python Code
+tag = "users"
+
+filtered_action_enums = toolset.find_actions_by_tags(
+    App.GITHUB,
+    tags=[tag], 
+)
+
+tools = toolset.get_tools(actions=filtered_action_enums)
+```
+
+- Filtering tools based on use case:
+```python Code
+use_case = "Star a repository on GitHub"
+
+filtered_action_enums = toolset.find_actions_by_use_case(
+    App.GITHUB, use_case=use_case, advanced=False
+)
+
+tools = toolset.get_tools(actions=filtered_action_enums)
+```
+<Tip>Set `advanced` to True to get actions for complex use cases</Tip>
+
+- Using specific tools:
+
+In this demo, we will use the `GITHUB_STAR_A_REPOSITORY_FOR_THE_AUTHENTICATED_USER` action from the GitHub app.
+```python Code
+tools = toolset.get_tools(
+    actions=[Action.GITHUB_STAR_A_REPOSITORY_FOR_THE_AUTHENTICATED_USER]
+)
+```
+Learn more about filtering actions [here](https://docs.composio.dev/patterns/tools/use-tools/use-specific-actions)
+
+4. Define agent
 
 ```python Code
 crewai_agent = Agent(
-    role="Github Agent",
-    goal="You take action on Github using Github APIs",
-    backstory=(
-        "You are AI agent that is responsible for taking actions on Github "
-        "on users behalf. You need to take action on Github using Github APIs"
-    ),
+    role="GitHub Agent",
+    goal="You take action on GitHub using GitHub APIs",
+    backstory="You are AI agent that is responsible for taking actions on GitHub on behalf of users using GitHub APIs",
     verbose=True,
     tools=tools,
+    llm= # pass an llm
 )
 ```
 
-3. Execute task
+5. Execute task
 
 ```python Code
 task = Task(
-    description="Star a repo ComposioHQ/composio on GitHub",
+    description="Star a repo composiohq/composio on GitHub",
     agent=crewai_agent,
-    expected_output="if the star happened",
+    expected_output="Status of the operation",
 )
 
-task.execute()
+crew = Crew(agents=[crewai_agent], tasks=[task])
+
+crew.kickoff()
 ```
 
-* More detailed list of tools can be found [here](https://app.composio.dev)
+* More detailed list of tools can be found [here](https://app.composio.dev)

docs/tools/filewritetool.mdx

@@ -8,9 +8,9 @@ icon: file-pen
 
 ## Description
 
-The `FileWriterTool` is a component of the crewai_tools package, designed to simplify the process of writing content to files. 
+The `FileWriterTool` is a component of the crewai_tools package, designed to simplify the process of writing content to files with cross-platform compatibility (Windows, Linux, macOS). 
 It is particularly useful in scenarios such as generating reports, saving logs, creating configuration files, and more. 
-This tool supports creating new directories if they don't exist, making it easier to organize your output.
+This tool handles path differences across operating systems, supports UTF-8 encoding, and automatically creates directories if they don't exist, making it easier to organize your output reliably across different platforms.
 
 ## Installation
 
@@ -43,6 +43,8 @@ print(result)
 
 ## Conclusion
 
-By integrating the `FileWriterTool` into your crews, the agents can execute the process of writing content to files and creating directories. 
-This tool is essential for tasks that require saving output data, creating structured file systems, and more. By adhering to the setup and usage guidelines provided, 
-incorporating this tool into projects is straightforward and efficient.
+By integrating the `FileWriterTool` into your crews, the agents can reliably write content to files across different operating systems. 
+This tool is essential for tasks that require saving output data, creating structured file systems, and handling cross-platform file operations. 
+It's particularly recommended for Windows users who may encounter file writing issues with standard Python file operations.
+
+By adhering to the setup and usage guidelines provided, incorporating this tool into projects is straightforward and ensures consistent file writing behavior across all platforms.

mkdocs.yml

@@ -152,6 +152,7 @@ nav:
     - Agent Monitoring with AgentOps: 'how-to/AgentOps-Observability.md'
     - Agent Monitoring with LangTrace: 'how-to/Langtrace-Observability.md'
     - Agent Monitoring with OpenLIT: 'how-to/openlit-Observability.md'
+    - Agent Monitoring with MLflow: 'how-to/mlflow-Observability.md'
   - Tools Docs:
     - Browserbase Web Loader: 'tools/BrowserbaseLoadTool.md'
     - Code Docs RAG Search: 'tools/CodeDocsSearchTool.md'

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "crewai"
-version = "0.95.0"
+version = "0.100.1"
 description = "Cutting-edge framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks."
 readme = "README.md"
 requires-python = ">=3.10,<3.13"
@@ -11,27 +11,22 @@ dependencies = [
     # Core Dependencies
     "pydantic>=2.4.2",
     "openai>=1.13.3",
-    "litellm==1.57.4",
+    "litellm==1.60.2",
     "instructor>=1.3.3",
-
     # Text Processing
     "pdfplumber>=0.11.4",
     "regex>=2024.9.11",
-
     # Telemetry and Monitoring
     "opentelemetry-api>=1.22.0",
     "opentelemetry-sdk>=1.22.0",
     "opentelemetry-exporter-otlp-proto-http>=1.22.0",
-
     # Data Handling
     "chromadb>=0.5.23",
     "openpyxl>=3.1.5",
     "pyvis>=0.3.2",
-
     # Authentication and Security
     "auth0-python>=4.7.1",
     "python-dotenv>=1.0.0",
-
     # Configuration and Utils
     "click>=8.1.7",
     "appdirs>=1.4.4",
@@ -40,7 +35,8 @@ dependencies = [
     "uv>=0.4.25",
     "tomli-w>=1.1.0",
     "tomli>=2.0.2",
-    "blinker>=1.9.0"
+    "blinker>=1.9.0",
+    "json5>=0.10.0",
 ]
 
 [project.urls]
@@ -49,7 +45,7 @@ Documentation = "https://docs.crewai.com"
 Repository = "https://github.com/crewAIInc/crewAI"
 
 [project.optional-dependencies]
-tools = ["crewai-tools>=0.25.5"]
+tools = ["crewai-tools>=0.32.1"]
 embeddings = [
     "tiktoken~=0.7.0"
 ]

src/crewai/__init__.py

@@ -14,7 +14,7 @@ warnings.filterwarnings(
     category=UserWarning,
     module="pydantic.main",
 )
-__version__ = "0.95.0"
+__version__ = "0.100.1"
 __all__ = [
     "Agent",
     "Crew",

src/crewai/agent.py

@@ -21,7 +21,6 @@ from crewai.tools.base_tool import Tool
 from crewai.utilities import Converter, Prompts
 from crewai.utilities.constants import TRAINED_AGENTS_DATA_FILE, TRAINING_DATA_FILE
 from crewai.utilities.converter import generate_model_description
-from crewai.utilities.llm_utils import create_llm
 from crewai.utilities.token_counter_callback import TokenCalcHandler
 from crewai.utilities.training_handler import CrewTrainingHandler
 
@@ -86,7 +85,7 @@ class Agent(BaseAgent):
     llm: Union[str, InstanceOf[LLM], Any] = Field(
         description="Language model that will run the agent.", default=None
     )
-    function_calling_llm: Optional[Union[str, InstanceOf[LLM], Any]] = Field(
+    function_calling_llm: Optional[Any] = Field(
         description="Language model that will run the agent.", default=None
     )
     system_template: Optional[str] = Field(
@@ -140,10 +139,89 @@ class Agent(BaseAgent):
     def post_init_setup(self):
         self._set_knowledge()
         self.agent_ops_agent_name = self.role
+        unaccepted_attributes = [
+            "AWS_ACCESS_KEY_ID",
+            "AWS_SECRET_ACCESS_KEY",
+            "AWS_REGION_NAME",
+        ]
 
-        self.llm = create_llm(self.llm)
-        if self.function_calling_llm and not isinstance(self.function_calling_llm, LLM):
-            self.function_calling_llm = create_llm(self.function_calling_llm)
+        # Handle different cases for self.llm
+        if isinstance(self.llm, str):
+            # If it's a string, create an LLM instance
+            self.llm = LLM(model=self.llm)
+        elif isinstance(self.llm, LLM):
+            # If it's already an LLM instance, keep it as is
+            pass
+        elif self.llm is None:
+            # Determine the model name from environment variables or use default
+            model_name = (
+                os.environ.get("OPENAI_MODEL_NAME")
+                or os.environ.get("MODEL")
+                or "gpt-4o-mini"
+            )
+            llm_params = {"model": model_name}
+
+            api_base = os.environ.get("OPENAI_API_BASE") or os.environ.get(
+                "OPENAI_BASE_URL"
+            )
+            if api_base:
+                llm_params["base_url"] = api_base
+
+            set_provider = model_name.split("/")[0] if "/" in model_name else "openai"
+
+            # Iterate over all environment variables to find matching API keys or use defaults
+            for provider, env_vars in ENV_VARS.items():
+                if provider == set_provider:
+                    for env_var in env_vars:
+                        # Check if the environment variable is set
+                        key_name = env_var.get("key_name")
+                        if key_name and key_name not in unaccepted_attributes:
+                            env_value = os.environ.get(key_name)
+                            if env_value:
+                                key_name = key_name.lower()
+                                for pattern in LITELLM_PARAMS:
+                                    if pattern in key_name:
+                                        key_name = pattern
+                                        break
+                                llm_params[key_name] = env_value
+                        # Check for default values if the environment variable is not set
+                        elif env_var.get("default", False):
+                            for key, value in env_var.items():
+                                if key not in ["prompt", "key_name", "default"]:
+                                    # Only add default if the key is already set in os.environ
+                                    if key in os.environ:
+                                        llm_params[key] = value
+
+            self.llm = LLM(**llm_params)
+        else:
+            # For any other type, attempt to extract relevant attributes
+            llm_params = {
+                "model": getattr(self.llm, "model_name", None)
+                or getattr(self.llm, "deployment_name", None)
+                or str(self.llm),
+                "temperature": getattr(self.llm, "temperature", None),
+                "max_tokens": getattr(self.llm, "max_tokens", None),
+                "logprobs": getattr(self.llm, "logprobs", None),
+                "timeout": getattr(self.llm, "timeout", None),
+                "max_retries": getattr(self.llm, "max_retries", None),
+                "api_key": getattr(self.llm, "api_key", None),
+                "base_url": getattr(self.llm, "base_url", None),
+                "organization": getattr(self.llm, "organization", None),
+            }
+            # Remove None values to avoid passing unnecessary parameters
+            llm_params = {k: v for k, v in llm_params.items() if v is not None}
+            self.llm = LLM(**llm_params)
+
+        # Similar handling for function_calling_llm
+        if self.function_calling_llm:
+            if isinstance(self.function_calling_llm, str):
+                self.function_calling_llm = LLM(model=self.function_calling_llm)
+            elif not isinstance(self.function_calling_llm, LLM):
+                self.function_calling_llm = LLM(
+                    model=getattr(self.function_calling_llm, "model_name", None)
+                    or getattr(self.function_calling_llm, "deployment_name", None)
+                    or str(self.function_calling_llm)
+                )
 
         if not self.agent_executor:
             self._setup_agent_executor()
@@ -165,6 +243,15 @@ class Agent(BaseAgent):
                 if isinstance(self.knowledge_sources, list) and all(
                     isinstance(k, BaseKnowledgeSource) for k in self.knowledge_sources
                 ):
+                    # Validate embedding configuration based on provider
+                    from crewai.utilities.constants import DEFAULT_EMBEDDING_PROVIDER
+                    provider = os.getenv("CREWAI_EMBEDDING_PROVIDER", DEFAULT_EMBEDDING_PROVIDER)
+                    
+                    if provider == "openai" and not os.getenv("OPENAI_API_KEY"):
+                        raise ValueError("Please provide an OpenAI API key via OPENAI_API_KEY environment variable")
+                    elif provider == "ollama" and not os.getenv("CREWAI_OLLAMA_URL", "http://localhost:11434/api/embeddings"):
+                        raise ValueError("Please provide Ollama URL via CREWAI_OLLAMA_URL environment variable")
+                    
                     self._knowledge = Knowledge(
                         sources=self.knowledge_sources,
                         embedder_config=self.embedder_config,
@@ -335,7 +422,6 @@ class Agent(BaseAgent):
 
     def get_multimodal_tools(self) -> List[Tool]:
         from crewai.tools.agent_tools.add_image_tool import AddImageTool
-
         return [AddImageTool()]
 
     def get_code_execution_tools(self):

src/crewai/agents/agent_builder/base_agent.py

@@ -18,10 +18,13 @@ from pydantic_core import PydanticCustomError
 from crewai.agents.agent_builder.utilities.base_token_process import TokenProcess
 from crewai.agents.cache.cache_handler import CacheHandler
 from crewai.agents.tools_handler import ToolsHandler
+from crewai.knowledge.knowledge import Knowledge
+from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
 from crewai.tools import BaseTool
 from crewai.tools.base_tool import Tool
 from crewai.utilities import I18N, Logger, RPMController
 from crewai.utilities.config import process_config
+from crewai.utilities.converter import Converter
 
 T = TypeVar("T", bound="BaseAgent")
 
@@ -40,7 +43,7 @@ class BaseAgent(ABC, BaseModel):
         max_rpm (Optional[int]): Maximum number of requests per minute for the agent execution.
         allow_delegation (bool): Allow delegation of tasks to agents.
         tools (Optional[List[Any]]): Tools at the agent's disposal.
-        max_iter (Optional[int]): Maximum iterations for an agent to execute a task.
+        max_iter (int): Maximum iterations for an agent to execute a task.
         agent_executor (InstanceOf): An instance of the CrewAgentExecutor class.
         llm (Any): Language model that will run the agent.
         crew (Any): Crew to which the agent belongs.
@@ -48,6 +51,8 @@ class BaseAgent(ABC, BaseModel):
         cache_handler (InstanceOf[CacheHandler]): An instance of the CacheHandler class.
         tools_handler (InstanceOf[ToolsHandler]): An instance of the ToolsHandler class.
         max_tokens: Maximum number of tokens for the agent to generate in a response.
+        knowledge_sources: Knowledge sources for the agent.
+        knowledge_storage: Custom knowledge storage for the agent.
 
 
     Methods:
@@ -108,9 +113,9 @@ class BaseAgent(ABC, BaseModel):
         description="Enable agent to delegate and ask questions among each other.",
     )
     tools: Optional[List[Any]] = Field(
-        default_factory=list, description="Tools at agents' disposal"
+        default_factory=lambda: [], description="Tools at agents' disposal"
     )
-    max_iter: Optional[int] = Field(
+    max_iter: int = Field(
         default=25, description="Maximum iterations for an agent to execute a task"
     )
     agent_executor: InstanceOf = Field(
@@ -121,15 +126,27 @@ class BaseAgent(ABC, BaseModel):
     )
     crew: Any = Field(default=None, description="Crew to which the agent belongs.")
     i18n: I18N = Field(default=I18N(), description="Internationalization settings.")
-    cache_handler: InstanceOf[CacheHandler] = Field(
+    cache_handler: Optional[InstanceOf[CacheHandler]] = Field(
         default=None, description="An instance of the CacheHandler class."
     )
     tools_handler: InstanceOf[ToolsHandler] = Field(
-        default=None, description="An instance of the ToolsHandler class."
+        default_factory=ToolsHandler,
+        description="An instance of the ToolsHandler class.",
     )
     max_tokens: Optional[int] = Field(
         default=None, description="Maximum number of tokens for the agent's execution."
     )
+    knowledge: Optional[Knowledge] = Field(
+        default=None, description="Knowledge for the agent."
+    )
+    knowledge_sources: Optional[List[BaseKnowledgeSource]] = Field(
+        default=None,
+        description="Knowledge sources for the agent.",
+    )
+    knowledge_storage: Optional[Any] = Field(
+        default=None,
+        description="Custom knowledge storage for the agent.",
+    )
 
     @model_validator(mode="before")
     @classmethod
@@ -239,7 +256,7 @@ class BaseAgent(ABC, BaseModel):
     @abstractmethod
     def get_output_converter(
         self, llm: Any, text: str, model: type[BaseModel] | None, instructions: str
-    ):
+    ) -> Converter:
         """Get the converter class for the agent to create json/pydantic outputs."""
         pass
 
@@ -256,13 +273,44 @@ class BaseAgent(ABC, BaseModel):
             "tools_handler",
             "cache_handler",
             "llm",
+            "knowledge_sources",
+            "knowledge_storage",
+            "knowledge",
         }
 
-        # Copy llm and clear callbacks
+        # Copy llm
         existing_llm = shallow_copy(self.llm)
+        copied_knowledge = shallow_copy(self.knowledge)
+        copied_knowledge_storage = shallow_copy(self.knowledge_storage)
+        # Properly copy knowledge sources if they exist
+        existing_knowledge_sources = None
+        if self.knowledge_sources:
+            # Create a shared storage instance for all knowledge sources
+            shared_storage = (
+                self.knowledge_sources[0].storage if self.knowledge_sources else None
+            )
+
+            existing_knowledge_sources = []
+            for source in self.knowledge_sources:
+                copied_source = (
+                    source.model_copy()
+                    if hasattr(source, "model_copy")
+                    else shallow_copy(source)
+                )
+                # Ensure all copied sources use the same storage instance
+                copied_source.storage = shared_storage
+                existing_knowledge_sources.append(copied_source)
+
         copied_data = self.model_dump(exclude=exclude)
         copied_data = {k: v for k, v in copied_data.items() if v is not None}
-        copied_agent = type(self)(**copied_data, llm=existing_llm, tools=self.tools)
+        copied_agent = type(self)(
+            **copied_data,
+            llm=existing_llm,
+            tools=self.tools,
+            knowledge_sources=existing_knowledge_sources,
+            knowledge=copied_knowledge,
+            knowledge_storage=copied_knowledge_storage,
+        )
 
         return copied_agent
 

src/crewai/agents/agent_builder/base_agent_executor_mixin.py

@@ -95,18 +95,29 @@ class CrewAgentExecutorMixin:
                 pass
 
     def _ask_human_input(self, final_answer: str) -> str:
-        """Prompt human input for final decision making."""
+        """Prompt human input with mode-appropriate messaging."""
         self._printer.print(
             content=f"\033[1m\033[95m ## Final Result:\033[00m \033[92m{final_answer}\033[00m"
         )
 
-        self._printer.print(
-            content=(
+        # Training mode prompt (single iteration)
+        if self.crew and getattr(self.crew, "_train", False):
+            prompt = (
                 "\n\n=====\n"
-                "## Please provide feedback on the Final Result and the Agent's actions. "
-                "Respond with 'looks good' or a similar phrase when you're satisfied.\n"
+                "## TRAINING MODE: Provide feedback to improve the agent's performance.\n"
+                "This will be used to train better versions of the agent.\n"
+                "Please provide detailed feedback about the result quality and reasoning process.\n"
                 "=====\n"
-            ),
-            color="bold_yellow",
-        )
+            )
+        # Regular human-in-the-loop prompt (multiple iterations)
+        else:
+            prompt = (
+                "\n\n=====\n"
+                "## HUMAN FEEDBACK: Provide feedback on the Final Result and Agent's actions.\n"
+                "Respond with 'looks good' to accept or provide specific improvement requests.\n"
+                "You can provide multiple rounds of feedback until satisfied.\n"
+                "=====\n"
+            )
+
+        self._printer.print(content=prompt, color="bold_yellow")
         return input()

src/crewai/agents/agent_builder/utilities/base_output_converter.py

@@ -25,7 +25,7 @@ class OutputConverter(BaseModel, ABC):
     llm: Any = Field(description="The language model to be used to convert the text.")
     model: Any = Field(description="The model to be used to convert the text.")
     instructions: str = Field(description="Conversion instructions to the LLM.")
-    max_attempts: Optional[int] = Field(
+    max_attempts: int = Field(
         description="Max number of attempts to try to get the output formatted.",
         default=3,
     )

src/crewai/agents/agent_builder/utilities/base_token_process.py

@@ -2,25 +2,26 @@ from crewai.types.usage_metrics import UsageMetrics
 
 
 class TokenProcess:
-    total_tokens: int = 0
-    prompt_tokens: int = 0
-    cached_prompt_tokens: int = 0
-    completion_tokens: int = 0
-    successful_requests: int = 0
+    def __init__(self) -> None:
+        self.total_tokens: int = 0
+        self.prompt_tokens: int = 0
+        self.cached_prompt_tokens: int = 0
+        self.completion_tokens: int = 0
+        self.successful_requests: int = 0
 
-    def sum_prompt_tokens(self, tokens: int):
-        self.prompt_tokens = self.prompt_tokens + tokens
-        self.total_tokens = self.total_tokens + tokens
+    def sum_prompt_tokens(self, tokens: int) -> None:
+        self.prompt_tokens += tokens
+        self.total_tokens += tokens
 
-    def sum_completion_tokens(self, tokens: int):
-        self.completion_tokens = self.completion_tokens + tokens
-        self.total_tokens = self.total_tokens + tokens
+    def sum_completion_tokens(self, tokens: int) -> None:
+        self.completion_tokens += tokens
+        self.total_tokens += tokens
 
-    def sum_cached_prompt_tokens(self, tokens: int):
-        self.cached_prompt_tokens = self.cached_prompt_tokens + tokens
+    def sum_cached_prompt_tokens(self, tokens: int) -> None:
+        self.cached_prompt_tokens += tokens
 
-    def sum_successful_requests(self, requests: int):
-        self.successful_requests = self.successful_requests + requests
+    def sum_successful_requests(self, requests: int) -> None:
+        self.successful_requests += requests
 
     def get_summary(self) -> UsageMetrics:
         return UsageMetrics(

src/crewai/agents/crew_agent_executor.py

@@ -13,6 +13,7 @@ from crewai.agents.parser import (
     OutputParserException,
 )
 from crewai.agents.tools_handler import ToolsHandler
+from crewai.llm import LLM
 from crewai.tools.base_tool import BaseTool
 from crewai.tools.tool_usage import ToolUsage, ToolUsageErrorException
 from crewai.utilities import I18N, Printer
@@ -54,7 +55,7 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         callbacks: List[Any] = [],
     ):
         self._i18n: I18N = I18N()
-        self.llm = llm
+        self.llm: LLM = llm
         self.task = task
         self.agent = agent
         self.crew = crew
@@ -80,10 +81,8 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         self.tool_name_to_tool_map: Dict[str, BaseTool] = {
             tool.name: tool for tool in self.tools
         }
-        if self.llm.stop:
-            self.llm.stop = list(set(self.llm.stop + self.stop))
-        else:
-            self.llm.stop = self.stop
+        self.stop = stop_words
+        self.llm.stop = list(set(self.llm.stop + self.stop))
 
     def invoke(self, inputs: Dict[str, str]) -> Dict[str, Any]:
         if "system" in self.prompt:
@@ -98,7 +97,22 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         self._show_start_logs()
 
         self.ask_for_human_input = bool(inputs.get("ask_for_human_input", False))
-        formatted_answer = self._invoke_loop()
+
+        try:
+            formatted_answer = self._invoke_loop()
+        except AssertionError:
+            self._printer.print(
+                content="Agent failed to reach a final answer. This is likely a bug - please report it.",
+                color="red",
+            )
+            raise
+        except Exception as e:
+            if e.__class__.__module__.startswith("litellm"):
+                # Do not retry on litellm errors
+                raise e
+            else:
+                self._handle_unknown_error(e)
+                raise e
 
         if self.ask_for_human_input:
             formatted_answer = self._handle_human_feedback(formatted_answer)
@@ -107,7 +121,7 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         self._create_long_term_memory(formatted_answer)
         return {"output": formatted_answer.output}
 
-    def _invoke_loop(self):
+    def _invoke_loop(self) -> AgentFinish:
         """
         Main loop to invoke the agent's thought process until it reaches a conclusion
         or the maximum number of iterations is reached.
@@ -124,7 +138,6 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
                 self._enforce_rpm_limit()
 
                 answer = self._get_llm_response()
-
                 formatted_answer = self._process_llm_response(answer)
 
                 if isinstance(formatted_answer, AgentAction):
@@ -142,13 +155,37 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
                 formatted_answer = self._handle_output_parser_exception(e)
 
             except Exception as e:
+                if e.__class__.__module__.startswith("litellm"):
+                    # Do not retry on litellm errors
+                    raise e
                 if self._is_context_length_exceeded(e):
                     self._handle_context_length()
                     continue
+                else:
+                    self._handle_unknown_error(e)
+                    raise e
+            finally:
+                self.iterations += 1
 
+        # During the invoke loop, formatted_answer alternates between AgentAction
+        # (when the agent is using tools) and eventually becomes AgentFinish
+        # (when the agent reaches a final answer). This assertion confirms we've
+        # reached a final answer and helps type checking understand this transition.
+        assert isinstance(formatted_answer, AgentFinish)
         self._show_logs(formatted_answer)
         return formatted_answer
 
+    def _handle_unknown_error(self, exception: Exception) -> None:
+        """Handle unknown errors by informing the user."""
+        self._printer.print(
+            content="An unknown error occurred. Please check the details below.",
+            color="red",
+        )
+        self._printer.print(
+            content=f"Error details: {exception}",
+            color="red",
+        )
+
     def _has_reached_max_iterations(self) -> bool:
         """Check if the maximum number of iterations has been reached."""
         return self.iterations >= self.max_iter
@@ -160,10 +197,17 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
 
     def _get_llm_response(self) -> str:
         """Call the LLM and return the response, handling any invalid responses."""
-        answer = self.llm.call(
-            self.messages,
-            callbacks=self.callbacks,
-        )
+        try:
+            answer = self.llm.call(
+                self.messages,
+                callbacks=self.callbacks,
+            )
+        except Exception as e:
+            self._printer.print(
+                content=f"Error during LLM call: {e}",
+                color="red",
+            )
+            raise e
 
         if not answer:
             self._printer.print(
@@ -184,7 +228,6 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
                 if FINAL_ANSWER_AND_PARSABLE_ACTION_ERROR_MESSAGE in e.error:
                     answer = answer.split("Observation:")[0].strip()
 
-        self.iterations += 1
         return self._format_answer(answer)
 
     def _handle_agent_action(
@@ -260,8 +303,11 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
             self._printer.print(
                 content=f"\033[1m\033[95m# Agent:\033[00m \033[1m\033[92m{agent_role}\033[00m"
             )
+            description = (
+                getattr(self.task, "description") if self.task else "Not Found"
+            )
             self._printer.print(
-                content=f"\033[95m## Task:\033[00m \033[92m{self.task.description}\033[00m"
+                content=f"\033[95m## Task:\033[00m \033[92m{description}\033[00m"
             )
 
     def _show_logs(self, formatted_answer: Union[AgentAction, AgentFinish]):
@@ -386,58 +432,50 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
             )
 
     def _handle_crew_training_output(
-        self, result: AgentFinish, human_feedback: str | None = None
+        self, result: AgentFinish, human_feedback: Optional[str] = None
     ) -> None:
-        """Function to handle the process of the training data."""
+        """Handle the process of saving training data."""
         agent_id = str(self.agent.id)  # type: ignore
+        train_iteration = (
+            getattr(self.crew, "_train_iteration", None) if self.crew else None
+        )
+
+        if train_iteration is None or not isinstance(train_iteration, int):
+            self._printer.print(
+                content="Invalid or missing train iteration. Cannot save training data.",
+                color="red",
+            )
+            return
 
-        # Load training data
         training_handler = CrewTrainingHandler(TRAINING_DATA_FILE)
-        training_data = training_handler.load()
+        training_data = training_handler.load() or {}
 
-        # Check if training data exists, human input is not requested, and self.crew is valid
-        if training_data and not self.ask_for_human_input:
-            if self.crew is not None and hasattr(self.crew, "_train_iteration"):
-                train_iteration = self.crew._train_iteration
-                if agent_id in training_data and isinstance(train_iteration, int):
-                    training_data[agent_id][train_iteration][
-                        "improved_output"
-                    ] = result.output
-                    training_handler.save(training_data)
-                else:
-                    self._printer.print(
-                        content="Invalid train iteration type or agent_id not in training data.",
-                        color="red",
-                    )
-            else:
-                self._printer.print(
-                    content="Crew is None or does not have _train_iteration attribute.",
-                    color="red",
-                )
+        # Initialize or retrieve agent's training data
+        agent_training_data = training_data.get(agent_id, {})
 
-        if self.ask_for_human_input and human_feedback is not None:
-            training_data = {
+        if human_feedback is not None:
+            # Save initial output and human feedback
+            agent_training_data[train_iteration] = {
                 "initial_output": result.output,
                 "human_feedback": human_feedback,
-                "agent": agent_id,
-                "agent_role": self.agent.role,  # type: ignore
             }
-            if self.crew is not None and hasattr(self.crew, "_train_iteration"):
-                train_iteration = self.crew._train_iteration
-                if isinstance(train_iteration, int):
-                    CrewTrainingHandler(TRAINING_DATA_FILE).append(
-                        train_iteration, agent_id, training_data
-                    )
-                else:
-                    self._printer.print(
-                        content="Invalid train iteration type. Expected int.",
-                        color="red",
-                    )
+        else:
+            # Save improved output
+            if train_iteration in agent_training_data:
+                agent_training_data[train_iteration]["improved_output"] = result.output
             else:
                 self._printer.print(
-                    content="Crew is None or does not have _train_iteration attribute.",
+                    content=(
+                        f"No existing training data for agent {agent_id} and iteration "
+                        f"{train_iteration}. Cannot save improved output."
+                    ),
                     color="red",
                 )
+                return
+
+        # Update the training data and save
+        training_data[agent_id] = agent_training_data
+        training_handler.save(training_data)
 
     def _format_prompt(self, prompt: str, inputs: Dict[str, str]) -> str:
         prompt = prompt.replace("{input}", inputs["input"])
@@ -453,82 +491,111 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
         return {"role": role, "content": prompt}
 
     def _handle_human_feedback(self, formatted_answer: AgentFinish) -> AgentFinish:
-        """
-        Handles the human feedback loop, allowing the user to provide feedback
-        on the agent's output and determining if additional iterations are needed.
+        """Handle human feedback with different flows for training vs regular use.
 
-        Parameters:
-            formatted_answer (AgentFinish): The initial output from the agent.
+        Args:
+            formatted_answer: The initial AgentFinish result to get feedback on
 
         Returns:
-            AgentFinish: The final output after incorporating human feedback.
+            AgentFinish: The final answer after processing feedback
         """
+        human_feedback = self._ask_human_input(formatted_answer.output)
+
+        if self._is_training_mode():
+            return self._handle_training_feedback(formatted_answer, human_feedback)
+
+        return self._handle_regular_feedback(formatted_answer, human_feedback)
+
+    def _is_training_mode(self) -> bool:
+        """Check if crew is in training mode."""
+        return bool(self.crew and self.crew._train)
+
+    def _handle_training_feedback(
+        self, initial_answer: AgentFinish, feedback: str
+    ) -> AgentFinish:
+        """Process feedback for training scenarios with single iteration."""
+        self._printer.print(
+            content="\nProcessing training feedback.\n",
+            color="yellow",
+        )
+        self._handle_crew_training_output(initial_answer, feedback)
+        self.messages.append(
+            self._format_msg(
+                self._i18n.slice("feedback_instructions").format(feedback=feedback)
+            )
+        )
+        improved_answer = self._invoke_loop()
+        self._handle_crew_training_output(improved_answer)
+        self.ask_for_human_input = False
+        return improved_answer
+
+    def _handle_regular_feedback(
+        self, current_answer: AgentFinish, initial_feedback: str
+    ) -> AgentFinish:
+        """Process feedback for regular use with potential multiple iterations."""
+        feedback = initial_feedback
+        answer = current_answer
+
         while self.ask_for_human_input:
-            human_feedback = self._ask_human_input(formatted_answer.output)
+            response = self._get_llm_feedback_response(feedback)
 
-            if self.crew and self.crew._train:
-                self._handle_crew_training_output(formatted_answer, human_feedback)
-
-            # Make an LLM call to verify if additional changes are requested based on human feedback
-            additional_changes_prompt = self._i18n.slice(
-                "human_feedback_classification"
-            ).format(feedback=human_feedback)
-
-            retry_count = 0
-            llm_call_successful = False
-            additional_changes_response = None
-
-            while retry_count < MAX_LLM_RETRY and not llm_call_successful:
-                try:
-                    additional_changes_response = (
-                        self.llm.call(
-                            [
-                                self._format_msg(
-                                    additional_changes_prompt, role="system"
-                                )
-                            ],
-                            callbacks=self.callbacks,
-                        )
-                        .strip()
-                        .lower()
-                    )
-                    llm_call_successful = True
-                except Exception as e:
-                    retry_count += 1
-
-                    self._printer.print(
-                        content=f"Error during LLM call to classify human feedback: {e}. Retrying... ({retry_count}/{MAX_LLM_RETRY})",
-                        color="red",
-                    )
-
-            if not llm_call_successful:
-                self._printer.print(
-                    content="Error processing feedback after multiple attempts.",
-                    color="red",
-                )
+            if not self._feedback_requires_changes(response):
                 self.ask_for_human_input = False
-                break
-
-            if additional_changes_response == "false":
-                self.ask_for_human_input = False
-            elif additional_changes_response == "true":
-                self.ask_for_human_input = True
-                # Add human feedback to messages
-                self.messages.append(self._format_msg(f"Feedback: {human_feedback}"))
-                # Invoke the loop again with updated messages
-                formatted_answer = self._invoke_loop()
-
-                if self.crew and self.crew._train:
-                    self._handle_crew_training_output(formatted_answer)
             else:
-                # Unexpected response
-                self._printer.print(
-                    content=f"Unexpected response from LLM: '{additional_changes_response}'. Assuming no additional changes requested.",
-                    color="red",
-                )
-                self.ask_for_human_input = False
+                answer = self._process_feedback_iteration(feedback)
+                feedback = self._ask_human_input(answer.output)
 
-        return formatted_answer
+        return answer
+
+    def _get_llm_feedback_response(self, feedback: str) -> Optional[str]:
+        """Get LLM classification of whether feedback requires changes."""
+        prompt = self._i18n.slice("human_feedback_classification").format(
+            feedback=feedback
+        )
+        message = self._format_msg(prompt, role="system")
+
+        for retry in range(MAX_LLM_RETRY):
+            try:
+                response = self.llm.call([message], callbacks=self.callbacks)
+                return response.strip().lower() if response else None
+            except Exception as error:
+                self._log_feedback_error(retry, error)
+
+        self._log_max_retries_exceeded()
+        return None
+
+    def _feedback_requires_changes(self, response: Optional[str]) -> bool:
+        """Determine if feedback response indicates need for changes."""
+        return response == "true" if response else False
+
+    def _process_feedback_iteration(self, feedback: str) -> AgentFinish:
+        """Process a single feedback iteration."""
+        self.messages.append(
+            self._format_msg(
+                self._i18n.slice("feedback_instructions").format(feedback=feedback)
+            )
+        )
+        return self._invoke_loop()
+
+    def _log_feedback_error(self, retry_count: int, error: Exception) -> None:
+        """Log feedback processing errors."""
+        self._printer.print(
+            content=(
+                f"Error processing feedback: {error}. "
+                f"Retrying... ({retry_count + 1}/{MAX_LLM_RETRY})"
+            ),
+            color="red",
+        )
+
+    def _log_max_retries_exceeded(self) -> None:
+        """Log when max retries for feedback processing are exceeded."""
+        self._printer.print(
+            content=(
+                f"Failed to process feedback after {MAX_LLM_RETRY} attempts. "
+                "Ending feedback loop."
+            ),
+            color="red",
+        )
 
     def _handle_max_iterations_exceeded(self, formatted_answer):
         """

src/crewai/cli/cli.py

@@ -350,7 +350,10 @@ def chat():
     Start a conversation with the Crew, collecting user-supplied inputs,
     and using the Chat LLM to generate responses.
     """
-    click.echo("Starting a conversation with the Crew")
+    click.secho(
+        "\nStarting a conversation with the Crew\n" "Type 'exit' or Ctrl+C to quit.\n",
+    )
+
     run_chat()
 
 

src/crewai/cli/crew_chat.py

@@ -1,17 +1,52 @@
 import json
+import platform
 import re
 import sys
+import threading
+import time
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Set, Tuple
 
 import click
 import tomli
+from packaging import version
 
+from crewai.cli.utils import read_toml
+from crewai.cli.version import get_crewai_version
 from crewai.crew import Crew
 from crewai.llm import LLM
 from crewai.types.crew_chat import ChatInputField, ChatInputs
 from crewai.utilities.llm_utils import create_llm
 
+MIN_REQUIRED_VERSION = "0.98.0"
+
+
+def check_conversational_crews_version(
+    crewai_version: str, pyproject_data: dict
+) -> bool:
+    """
+    Check if the installed crewAI version supports conversational crews.
+
+    Args:
+        crewai_version: The current version of crewAI.
+        pyproject_data: Dictionary containing pyproject.toml data.
+
+    Returns:
+        bool: True if version check passes, False otherwise.
+    """
+    try:
+        if version.parse(crewai_version) < version.parse(MIN_REQUIRED_VERSION):
+            click.secho(
+                "You are using an older version of crewAI that doesn't support conversational crews. "
+                "Run 'uv upgrade crewai' to get the latest version.",
+                fg="red",
+            )
+            return False
+    except version.InvalidVersion:
+        click.secho("Invalid crewAI version format detected.", fg="red")
+        return False
+    return True
+
 
 def run_chat():
     """
@@ -19,20 +54,47 @@ def run_chat():
     Incorporates crew_name, crew_description, and input fields to build a tool schema.
     Exits if crew_name or crew_description are missing.
     """
+    crewai_version = get_crewai_version()
+    pyproject_data = read_toml()
+
+    if not check_conversational_crews_version(crewai_version, pyproject_data):
+        return
+
     crew, crew_name = load_crew_and_name()
     chat_llm = initialize_chat_llm(crew)
     if not chat_llm:
         return
 
-    crew_chat_inputs = generate_crew_chat_inputs(crew, crew_name, chat_llm)
-    crew_tool_schema = generate_crew_tool_schema(crew_chat_inputs)
-    system_message = build_system_message(crew_chat_inputs)
-
-    # Call the LLM to generate the introductory message
-    introductory_message = chat_llm.call(
-        messages=[{"role": "system", "content": system_message}]
+    # Indicate that the crew is being analyzed
+    click.secho(
+        "\nAnalyzing crew and required inputs - this may take 3 to 30 seconds "
+        "depending on the complexity of your crew.",
+        fg="white",
     )
-    click.secho(f"\nAssistant: {introductory_message}\n", fg="green")
+
+    # Start loading indicator
+    loading_complete = threading.Event()
+    loading_thread = threading.Thread(target=show_loading, args=(loading_complete,))
+    loading_thread.start()
+
+    try:
+        crew_chat_inputs = generate_crew_chat_inputs(crew, crew_name, chat_llm)
+        crew_tool_schema = generate_crew_tool_schema(crew_chat_inputs)
+        system_message = build_system_message(crew_chat_inputs)
+
+        # Call the LLM to generate the introductory message
+        introductory_message = chat_llm.call(
+            messages=[{"role": "system", "content": system_message}]
+        )
+    finally:
+        # Stop loading indicator
+        loading_complete.set()
+        loading_thread.join()
+
+    # Indicate that the analysis is complete
+    click.secho("\nFinished analyzing crew.\n", fg="white")
+
+    click.secho(f"Assistant: {introductory_message}\n", fg="green")
 
     messages = [
         {"role": "system", "content": system_message},
@@ -43,15 +105,17 @@ def run_chat():
         crew_chat_inputs.crew_name: create_tool_function(crew, messages),
     }
 
-    click.secho(
-        "\nEntering an interactive chat loop with function-calling.\n"
-        "Type 'exit' or Ctrl+C to quit.\n",
-        fg="cyan",
-    )
-
     chat_loop(chat_llm, messages, crew_tool_schema, available_functions)
 
 
+def show_loading(event: threading.Event):
+    """Display animated loading dots while processing."""
+    while not event.is_set():
+        print(".", end="", flush=True)
+        time.sleep(1)
+    print()
+
+
 def initialize_chat_llm(crew: Crew) -> Optional[LLM]:
     """Initializes the chat LLM and handles exceptions."""
     try:
@@ -85,7 +149,7 @@ def build_system_message(crew_chat_inputs: ChatInputs) -> str:
         "Please keep your responses concise and friendly. "
         "If a user asks a question outside the crew's scope, provide a brief answer and remind them of the crew's purpose. "
         "After calling the tool, be prepared to take user feedback and make adjustments as needed. "
-        "If you are ever unsure about a user's request or need clarification, ask the user for more information."
+        "If you are ever unsure about a user's request or need clarification, ask the user for more information. "
         "Before doing anything else, introduce yourself with a friendly message like: 'Hey! I'm here to help you with [crew's purpose]. Could you please provide me with [inputs] so we can get started?' "
         "For example: 'Hey! I'm here to help you with uncovering and reporting cutting-edge developments through thorough research and detailed analysis. Could you please provide me with a topic you're interested in? This will help us generate a comprehensive research report and detailed analysis.'"
         f"\nCrew Name: {crew_chat_inputs.crew_name}"
@@ -102,25 +166,33 @@ def create_tool_function(crew: Crew, messages: List[Dict[str, str]]) -> Any:
     return run_crew_tool_with_messages
 
 
+def flush_input():
+    """Flush any pending input from the user."""
+    if platform.system() == "Windows":
+        # Windows platform
+        import msvcrt
+
+        while msvcrt.kbhit():
+            msvcrt.getch()
+    else:
+        # Unix-like platforms (Linux, macOS)
+        import termios
+
+        termios.tcflush(sys.stdin, termios.TCIFLUSH)
+
+
 def chat_loop(chat_llm, messages, crew_tool_schema, available_functions):
     """Main chat loop for interacting with the user."""
     while True:
         try:
-            user_input = click.prompt("You", type=str)
-            if user_input.strip().lower() in ["exit", "quit"]:
-                click.echo("Exiting chat. Goodbye!")
-                break
+            # Flush any pending input before accepting new input
+            flush_input()
 
-            messages.append({"role": "user", "content": user_input})
-            final_response = chat_llm.call(
-                messages=messages,
-                tools=[crew_tool_schema],
-                available_functions=available_functions,
+            user_input = get_user_input()
+            handle_user_input(
+                user_input, chat_llm, messages, crew_tool_schema, available_functions
             )
 
-            messages.append({"role": "assistant", "content": final_response})
-            click.secho(f"\nAssistant: {final_response}\n", fg="green")
-
         except KeyboardInterrupt:
             click.echo("\nExiting chat. Goodbye!")
             break
@@ -129,6 +201,55 @@ def chat_loop(chat_llm, messages, crew_tool_schema, available_functions):
             break
 
 
+def get_user_input() -> str:
+    """Collect multi-line user input with exit handling."""
+    click.secho(
+        "\nYou (type your message below. Press 'Enter' twice when you're done):",
+        fg="blue",
+    )
+    user_input_lines = []
+    while True:
+        line = input()
+        if line.strip().lower() == "exit":
+            return "exit"
+        if line == "":
+            break
+        user_input_lines.append(line)
+    return "\n".join(user_input_lines)
+
+
+def handle_user_input(
+    user_input: str,
+    chat_llm: LLM,
+    messages: List[Dict[str, str]],
+    crew_tool_schema: Dict[str, Any],
+    available_functions: Dict[str, Any],
+) -> None:
+    if user_input.strip().lower() == "exit":
+        click.echo("Exiting chat. Goodbye!")
+        return
+
+    if not user_input.strip():
+        click.echo("Empty message. Please provide input or type 'exit' to quit.")
+        return
+
+    messages.append({"role": "user", "content": user_input})
+
+    # Indicate that assistant is processing
+    click.echo()
+    click.secho("Assistant is processing your input. Please wait...", fg="green")
+
+    # Process assistant's response
+    final_response = chat_llm.call(
+        messages=messages,
+        tools=[crew_tool_schema],
+        available_functions=available_functions,
+    )
+
+    messages.append({"role": "assistant", "content": final_response})
+    click.secho(f"\nAssistant: {final_response}\n", fg="green")
+
+
 def generate_crew_tool_schema(crew_inputs: ChatInputs) -> dict:
     """
     Dynamically build a Littellm 'function' schema for the given crew.
@@ -323,10 +444,10 @@ def generate_input_description_with_ai(input_name: str, crew: Crew, chat_llm) ->
         ):
             # Replace placeholders with input names
             task_description = placeholder_pattern.sub(
-                lambda m: m.group(1), task.description
+                lambda m: m.group(1), task.description or ""
             )
             expected_output = placeholder_pattern.sub(
-                lambda m: m.group(1), task.expected_output
+                lambda m: m.group(1), task.expected_output or ""
             )
             context_texts.append(f"Task Description: {task_description}")
             context_texts.append(f"Expected Output: {expected_output}")
@@ -337,10 +458,10 @@ def generate_input_description_with_ai(input_name: str, crew: Crew, chat_llm) ->
             or f"{{{input_name}}}" in agent.backstory
         ):
             # Replace placeholders with input names
-            agent_role = placeholder_pattern.sub(lambda m: m.group(1), agent.role)
-            agent_goal = placeholder_pattern.sub(lambda m: m.group(1), agent.goal)
+            agent_role = placeholder_pattern.sub(lambda m: m.group(1), agent.role or "")
+            agent_goal = placeholder_pattern.sub(lambda m: m.group(1), agent.goal or "")
             agent_backstory = placeholder_pattern.sub(
-                lambda m: m.group(1), agent.backstory
+                lambda m: m.group(1), agent.backstory or ""
             )
             context_texts.append(f"Agent Role: {agent_role}")
             context_texts.append(f"Agent Goal: {agent_goal}")
@@ -381,18 +502,20 @@ def generate_crew_description_with_ai(crew: Crew, chat_llm) -> str:
     for task in crew.tasks:
         # Replace placeholders with input names
         task_description = placeholder_pattern.sub(
-            lambda m: m.group(1), task.description
+            lambda m: m.group(1), task.description or ""
         )
         expected_output = placeholder_pattern.sub(
-            lambda m: m.group(1), task.expected_output
+            lambda m: m.group(1), task.expected_output or ""
         )
         context_texts.append(f"Task Description: {task_description}")
         context_texts.append(f"Expected Output: {expected_output}")
     for agent in crew.agents:
         # Replace placeholders with input names
-        agent_role = placeholder_pattern.sub(lambda m: m.group(1), agent.role)
-        agent_goal = placeholder_pattern.sub(lambda m: m.group(1), agent.goal)
-        agent_backstory = placeholder_pattern.sub(lambda m: m.group(1), agent.backstory)
+        agent_role = placeholder_pattern.sub(lambda m: m.group(1), agent.role or "")
+        agent_goal = placeholder_pattern.sub(lambda m: m.group(1), agent.goal or "")
+        agent_backstory = placeholder_pattern.sub(
+            lambda m: m.group(1), agent.backstory or ""
+        )
         context_texts.append(f"Agent Role: {agent_role}")
         context_texts.append(f"Agent Goal: {agent_goal}")
         context_texts.append(f"Agent Backstory: {agent_backstory}")

src/crewai/cli/reset_memories_command.py

@@ -2,11 +2,7 @@ import subprocess
 
 import click
 
-from crewai.knowledge.storage.knowledge_storage import KnowledgeStorage
-from crewai.memory.entity.entity_memory import EntityMemory
-from crewai.memory.long_term.long_term_memory import LongTermMemory
-from crewai.memory.short_term.short_term_memory import ShortTermMemory
-from crewai.utilities.task_output_storage_handler import TaskOutputStorageHandler
+from crewai.cli.utils import get_crew
 
 
 def reset_memories_command(
@@ -30,30 +26,35 @@ def reset_memories_command(
     """
 
     try:
+        crew = get_crew()
+        if not crew:
+            raise ValueError("No crew found.")
         if all:
-            ShortTermMemory().reset()
-            EntityMemory().reset()
-            LongTermMemory().reset()
-            TaskOutputStorageHandler().reset()
-            KnowledgeStorage().reset()
+            crew.reset_memories(command_type="all")
             click.echo("All memories have been reset.")
-        else:
-            if long:
-                LongTermMemory().reset()
-                click.echo("Long term memory has been reset.")
+            return
 
-            if short:
-                ShortTermMemory().reset()
-                click.echo("Short term memory has been reset.")
-            if entity:
-                EntityMemory().reset()
-                click.echo("Entity memory has been reset.")
-            if kickoff_outputs:
-                TaskOutputStorageHandler().reset()
-                click.echo("Latest Kickoff outputs stored has been reset.")
-            if knowledge:
-                KnowledgeStorage().reset()
-                click.echo("Knowledge has been reset.")
+        if not any([long, short, entity, kickoff_outputs, knowledge]):
+            click.echo(
+                "No memory type specified. Please specify at least one type to reset."
+            )
+            return
+
+        if long:
+            crew.reset_memories(command_type="long")
+            click.echo("Long term memory has been reset.")
+        if short:
+            crew.reset_memories(command_type="short")
+            click.echo("Short term memory has been reset.")
+        if entity:
+            crew.reset_memories(command_type="entity")
+            click.echo("Entity memory has been reset.")
+        if kickoff_outputs:
+            crew.reset_memories(command_type="kickoff_outputs")
+            click.echo("Latest Kickoff outputs stored has been reset.")
+        if knowledge:
+            crew.reset_memories(command_type="knowledge")
+            click.echo("Knowledge has been reset.")
 
     except subprocess.CalledProcessError as e:
         click.echo(f"An error occurred while resetting the memories: {e}", err=True)

src/crewai/cli/templates/crew/.gitignore

@@ -1,2 +1,3 @@
 .env
 __pycache__/
+.DS_Store

src/crewai/cli/templates/crew/pyproject.toml

@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
 authors = [{ name = "Your Name", email = "you@example.com" }]
 requires-python = ">=3.10,<3.13"
 dependencies = [
-    "crewai[tools]>=0.95.0,<1.0.0"
+    "crewai[tools]>=0.100.1,<1.0.0"
 ]
 
 [project.scripts]

src/crewai/cli/templates/flow/.gitignore

@@ -1,3 +1,4 @@
 .env
 __pycache__/
 lib/
+.DS_Store

src/crewai/cli/templates/flow/main.py

@@ -3,7 +3,7 @@ from random import randint
 
 from pydantic import BaseModel
 
-from crewai.flow.flow import Flow, listen, start
+from crewai.flow import Flow, listen, start
 
 from {{folder_name}}.crews.poem_crew.poem_crew import PoemCrew
 

src/crewai/cli/templates/flow/pyproject.toml

@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
 authors = [{ name = "Your Name", email = "you@example.com" }]
 requires-python = ">=3.10,<3.13"
 dependencies = [
-    "crewai[tools]>=0.95.0,<1.0.0",
+    "crewai[tools]>=0.100.1,<1.0.0",
 ]
 
 [project.scripts]

src/crewai/cli/templates/tool/pyproject.toml

@@ -5,7 +5,7 @@ description = "Power up your crews with {{folder_name}}"
 readme = "README.md"
 requires-python = ">=3.10,<3.13"
 dependencies = [
-    "crewai[tools]>=0.95.0"
+    "crewai[tools]>=0.100.1"
 ]
 
 [tool.crewai]

src/crewai/cli/utils.py

@@ -9,6 +9,7 @@ import tomli
 from rich.console import Console
 
 from crewai.cli.constants import ENV_VARS
+from crewai.crew import Crew
 
 if sys.version_info >= (3, 11):
     import tomllib
@@ -247,3 +248,64 @@ def write_env_file(folder_path, env_vars):
     with open(env_file_path, "w") as file:
         for key, value in env_vars.items():
             file.write(f"{key}={value}\n")
+
+
+def get_crew(crew_path: str = "crew.py", require: bool = False) -> Crew | None:
+    """Get the crew instance from the crew.py file."""
+    try:
+        import importlib.util
+        import os
+
+        for root, _, files in os.walk("."):
+            if "crew.py" in files:
+                crew_path = os.path.join(root, "crew.py")
+                try:
+                    spec = importlib.util.spec_from_file_location(
+                        "crew_module", crew_path
+                    )
+                    if not spec or not spec.loader:
+                        continue
+                    module = importlib.util.module_from_spec(spec)
+                    try:
+                        sys.modules[spec.name] = module
+                        spec.loader.exec_module(module)
+
+                        for attr_name in dir(module):
+                            attr = getattr(module, attr_name)
+                            try:
+                                if callable(attr) and hasattr(attr, "crew"):
+                                    crew_instance = attr().crew()
+                                    return crew_instance
+
+                            except Exception as e:
+                                print(f"Error processing attribute {attr_name}: {e}")
+                                continue
+
+                    except Exception as exec_error:
+                        print(f"Error executing module: {exec_error}")
+                        import traceback
+
+                        print(f"Traceback: {traceback.format_exc()}")
+
+                except (ImportError, AttributeError) as e:
+                    if require:
+                        console.print(
+                            f"Error importing crew from {crew_path}: {str(e)}",
+                            style="bold red",
+                        )
+                        continue
+
+                break
+
+        if require:
+            console.print("No valid Crew instance found in crew.py", style="bold red")
+            raise SystemExit
+        return None
+
+    except Exception as e:
+        if require:
+            console.print(
+                f"Unexpected error while loading crew: {str(e)}", style="bold red"
+            )
+            raise SystemExit
+        return None

src/crewai/crew.py

@@ -1,11 +1,10 @@
 import asyncio
 import json
-import re
 import uuid
 import warnings
 from concurrent.futures import Future
 from hashlib import md5
-from typing import Any, Callable, Dict, List, Optional, Set, Tuple, Union
+from typing import Any, Callable, Dict, List, Optional, Sequence, Tuple, Union
 
 from pydantic import (
     UUID4,
@@ -17,7 +16,6 @@ from pydantic import (
     field_validator,
     model_validator,
 )
-from pydantic_core import PydanticCustomError
 
 from crewai.agent import Agent
 from crewai.agents.agent_builder.base_agent import BaseAgent
@@ -37,7 +35,6 @@ from crewai.tasks.task_output import TaskOutput
 from crewai.telemetry import Telemetry
 from crewai.tools.agent_tools.agent_tools import AgentTools
 from crewai.tools.base_tool import Tool
-from crewai.types.crew_chat import ChatInputs
 from crewai.types.usage_metrics import UsageMetrics
 from crewai.utilities import I18N, FileHandler, Logger, RPMController
 from crewai.utilities.constants import TRAINING_DATA_FILE
@@ -84,6 +81,7 @@ class Crew(BaseModel):
         step_callback: Callback to be executed after each step for every agents execution.
         share_crew: Whether you want to share the complete crew information and execution with crewAI to make the library better, and allow us to train models.
         planning: Plan the crew execution and add the plan to the crew.
+        chat_llm: The language model used for orchestrating chat interactions with the crew.
     """
 
     __hash__ = object.__hash__  # type: ignore
@@ -182,9 +180,9 @@ class Crew(BaseModel):
         default=None,
         description="Path to the prompt json file to be used for the crew.",
     )
-    output_log_file: Optional[str] = Field(
+    output_log_file: Optional[Union[bool, str]] = Field(
         default=None,
-        description="output_log_file",
+        description="Path to the log file to be saved",
     )
     planning: Optional[bool] = Field(
         default=False,
@@ -210,8 +208,9 @@ class Crew(BaseModel):
         default=None,
         description="LLM used to handle chatting with the crew.",
     )
-    _knowledge: Optional[Knowledge] = PrivateAttr(
+    knowledge: Optional[Knowledge] = Field(
         default=None,
+        description="Knowledge for the crew.",
     )
 
     @field_validator("id", mode="before")
@@ -289,9 +288,9 @@ class Crew(BaseModel):
                 if isinstance(self.knowledge_sources, list) and all(
                     isinstance(k, BaseKnowledgeSource) for k in self.knowledge_sources
                 ):
-                    self._knowledge = Knowledge(
+                    self.knowledge = Knowledge(
                         sources=self.knowledge_sources,
-                        embedder_config=self.embedder,
+                        embedder=self.embedder,
                         collection_name="crew",
                     )
 
@@ -378,6 +377,22 @@ class Crew(BaseModel):
 
         return self
 
+    @model_validator(mode="after")
+    def validate_must_have_non_conditional_task(self) -> "Crew":
+        """Ensure that a crew has at least one non-conditional task."""
+        if not self.tasks:
+            return self
+        non_conditional_count = sum(
+            1 for task in self.tasks if not isinstance(task, ConditionalTask)
+        )
+        if non_conditional_count == 0:
+            raise PydanticCustomError(
+                "only_conditional_tasks",
+                "Crew must include at least one non-conditional task",
+                {},
+            )
+        return self
+
     @model_validator(mode="after")
     def validate_first_task(self) -> "Crew":
         """Ensure the first task is not a ConditionalTask."""
@@ -437,6 +452,8 @@ class Crew(BaseModel):
                         )
         return self
 
+
+
     @property
     def key(self) -> str:
         source = [agent.key for agent in self.agents] + [
@@ -492,21 +509,26 @@ class Crew(BaseModel):
         train_crew = self.copy()
         train_crew._setup_for_training(filename)
 
-        for n_iteration in range(n_iterations):
-            train_crew._train_iteration = n_iteration
-            train_crew.kickoff(inputs=inputs)
+        try:
+            for n_iteration in range(n_iterations):
+                train_crew._train_iteration = n_iteration
+                train_crew.kickoff(inputs=inputs)
 
-        training_data = CrewTrainingHandler(TRAINING_DATA_FILE).load()
+            training_data = CrewTrainingHandler(TRAINING_DATA_FILE).load()
 
-        for agent in train_crew.agents:
-            if training_data.get(str(agent.id)):
-                result = TaskEvaluator(agent).evaluate_training_data(
-                    training_data=training_data, agent_id=str(agent.id)
-                )
-
-                CrewTrainingHandler(filename).save_trained_data(
-                    agent_id=str(agent.role), trained_data=result.model_dump()
-                )
+            for agent in train_crew.agents:
+                if training_data.get(str(agent.id)):
+                    result = TaskEvaluator(agent).evaluate_training_data(
+                        training_data=training_data, agent_id=str(agent.id)
+                    )
+                    CrewTrainingHandler(filename).save_trained_data(
+                        agent_id=str(agent.role), trained_data=result.model_dump()
+                    )
+        except Exception as e:
+            self._logger.log("error", f"Training failed: {e}", color="red")
+            CrewTrainingHandler(TRAINING_DATA_FILE).clear()
+            CrewTrainingHandler(filename).clear()
+            raise
 
     def kickoff(
         self,
@@ -674,12 +696,7 @@ class Crew(BaseModel):
                 manager.tools = []
                 raise Exception("Manager agent should not have tools")
         else:
-            self.manager_llm = (
-                getattr(self.manager_llm, "model_name", None)
-                or getattr(self.manager_llm, "model", None)
-                or getattr(self.manager_llm, "deployment_name", None)
-                or self.manager_llm
-            )
+            self.manager_llm = create_llm(self.manager_llm)
             manager = Agent(
                 role=i18n.retrieve("hierarchical_manager_agent", "role"),
                 goal=i18n.retrieve("hierarchical_manager_agent", "goal"),
@@ -739,6 +756,7 @@ class Crew(BaseModel):
                     task, task_outputs, futures, task_index, was_replayed
                 )
                 if skipped_task_output:
+                    task_outputs.append(skipped_task_output)
                     continue
 
             if task.async_execution:
@@ -762,7 +780,7 @@ class Crew(BaseModel):
                     context=context,
                     tools=tools_for_task,
                 )
-                task_outputs = [task_output]
+                task_outputs.append(task_output)
                 self._process_task_result(task, task_output)
                 self._store_execution_log(task, task_output, task_index, was_replayed)
 
@@ -783,7 +801,7 @@ class Crew(BaseModel):
             task_outputs = self._process_async_tasks(futures, was_replayed)
             futures.clear()
 
-        previous_output = task_outputs[task_index - 1] if task_outputs else None
+        previous_output = task_outputs[-1] if task_outputs else None
         if previous_output is not None and not task.should_execute(previous_output):
             self._logger.log(
                 "debug",
@@ -905,11 +923,15 @@ class Crew(BaseModel):
             )
 
     def _create_crew_output(self, task_outputs: List[TaskOutput]) -> CrewOutput:
-        if len(task_outputs) != 1:
-            raise ValueError(
-                "Something went wrong. Kickoff should return only one task output."
-            )
-        final_task_output = task_outputs[0]
+        if not task_outputs:
+            raise ValueError("No task outputs available to create crew output.")
+            
+        # Filter out empty outputs and get the last valid one as the main output
+        valid_outputs = [t for t in task_outputs if t.raw]
+        if not valid_outputs:
+            raise ValueError("No valid task outputs available to create crew output.")
+        final_task_output = valid_outputs[-1]
+            
         final_string_output = final_task_output.raw
         self._finish_execution(final_string_output)
         token_usage = self.calculate_usage_metrics()
@@ -918,7 +940,7 @@ class Crew(BaseModel):
             raw=final_task_output.raw,
             pydantic=final_task_output.pydantic,
             json_dict=final_task_output.json_dict,
-            tasks_output=[task.output for task in self.tasks if task.output],
+            tasks_output=task_outputs,
             token_usage=token_usage,
         )
 
@@ -991,8 +1013,8 @@ class Crew(BaseModel):
         return result
 
     def query_knowledge(self, query: List[str]) -> Union[List[Dict[str, Any]], None]:
-        if self._knowledge:
-            return self._knowledge.query(query)
+        if self.knowledge:
+            return self.knowledge.query(query)
         return None
 
     def fetch_inputs(self) -> Set[str]:
@@ -1036,6 +1058,8 @@ class Crew(BaseModel):
             "_telemetry",
             "agents",
             "tasks",
+            "knowledge_sources",
+            "knowledge",
         }
 
         cloned_agents = [agent.copy() for agent in self.agents]
@@ -1043,6 +1067,9 @@ class Crew(BaseModel):
         task_mapping = {}
 
         cloned_tasks = []
+        existing_knowledge_sources = shallow_copy(self.knowledge_sources)
+        existing_knowledge = shallow_copy(self.knowledge)
+
         for task in self.tasks:
             cloned_task = task.copy(cloned_agents, task_mapping)
             cloned_tasks.append(cloned_task)
@@ -1062,7 +1089,13 @@ class Crew(BaseModel):
         copied_data.pop("agents", None)
         copied_data.pop("tasks", None)
 
-        copied_crew = Crew(**copied_data, agents=cloned_agents, tasks=cloned_tasks)
+        copied_crew = Crew(
+            **copied_data,
+            agents=cloned_agents,
+            tasks=cloned_tasks,
+            knowledge_sources=existing_knowledge_sources,
+            knowledge=existing_knowledge,
+        )
 
         return copied_crew
 
@@ -1134,3 +1167,80 @@ class Crew(BaseModel):
 
     def __repr__(self):
         return f"Crew(id={self.id}, process={self.process}, number_of_agents={len(self.agents)}, number_of_tasks={len(self.tasks)})"
+
+    def reset_memories(self, command_type: str) -> None:
+        """Reset specific or all memories for the crew.
+
+        Args:
+            command_type: Type of memory to reset.
+                Valid options: 'long', 'short', 'entity', 'knowledge',
+                'kickoff_outputs', or 'all'
+
+        Raises:
+            ValueError: If an invalid command type is provided.
+            RuntimeError: If memory reset operation fails.
+        """
+        VALID_TYPES = frozenset(
+            ["long", "short", "entity", "knowledge", "kickoff_outputs", "all"]
+        )
+
+        if command_type not in VALID_TYPES:
+            raise ValueError(
+                f"Invalid command type. Must be one of: {', '.join(sorted(VALID_TYPES))}"
+            )
+
+        try:
+            if command_type == "all":
+                self._reset_all_memories()
+            else:
+                self._reset_specific_memory(command_type)
+
+            self._logger.log("info", f"{command_type} memory has been reset")
+
+        except Exception as e:
+            error_msg = f"Failed to reset {command_type} memory: {str(e)}"
+            self._logger.log("error", error_msg)
+            raise RuntimeError(error_msg) from e
+
+    def _reset_all_memories(self) -> None:
+        """Reset all available memory systems."""
+        memory_systems = [
+            ("short term", self._short_term_memory),
+            ("entity", self._entity_memory),
+            ("long term", self._long_term_memory),
+            ("task output", self._task_output_handler),
+            ("knowledge", self.knowledge),
+        ]
+
+        for name, system in memory_systems:
+            if system is not None:
+                try:
+                    system.reset()
+                except Exception as e:
+                    raise RuntimeError(f"Failed to reset {name} memory") from e
+
+    def _reset_specific_memory(self, memory_type: str) -> None:
+        """Reset a specific memory system.
+
+        Args:
+            memory_type: Type of memory to reset
+
+        Raises:
+            RuntimeError: If the specified memory system fails to reset
+        """
+        reset_functions = {
+            "long": (self._long_term_memory, "long term"),
+            "short": (self._short_term_memory, "short term"),
+            "entity": (self._entity_memory, "entity"),
+            "knowledge": (self.knowledge, "knowledge"),
+            "kickoff_outputs": (self._task_output_handler, "task output"),
+        }
+
+        memory_system, name = reset_functions[memory_type]
+        if memory_system is None:
+            raise RuntimeError(f"{name} memory system is not initialized")
+
+        try:
+            memory_system.reset()
+        except Exception as e:
+            raise RuntimeError(f"Failed to reset {name} memory") from e

src/crewai/crews/crew_output.py

@@ -1,7 +1,9 @@
 import json
-from typing import Any, Dict, Optional
+from typing import Any, Callable, Dict, Optional
 
 from pydantic import BaseModel, Field
+from pydantic.main import IncEx
+from typing_extensions import Literal
 
 from crewai.tasks.output_format import OutputFormat
 from crewai.tasks.task_output import TaskOutput
@@ -21,16 +23,45 @@ class CrewOutput(BaseModel):
     tasks_output: list[TaskOutput] = Field(
         description="Output of each task", default=[]
     )
-    token_usage: UsageMetrics = Field(description="Processed token summary", default={})
+    token_usage: UsageMetrics = Field(description="Processed token summary", default_factory=UsageMetrics)
 
-    @property
-    def json(self) -> Optional[str]:
-        if self.tasks_output[-1].output_format != OutputFormat.JSON:
+    def model_json(self) -> str:
+        """Get the JSON representation of the output."""
+        if self.tasks_output and self.tasks_output[-1].output_format != OutputFormat.JSON:
             raise ValueError(
                 "No JSON output found in the final task. Please make sure to set the output_json property in the final task in your crew."
             )
+        return json.dumps(self.json_dict) if self.json_dict else "{}"
 
-        return json.dumps(self.json_dict)
+    def model_dump_json(
+        self,
+        *,
+        indent: Optional[int] = None,
+        include: Optional[IncEx] = None,
+        exclude: Optional[IncEx] = None,
+        context: Optional[Any] = None,
+        by_alias: bool = False,
+        exclude_unset: bool = False,
+        exclude_defaults: bool = False,
+        exclude_none: bool = False,
+        round_trip: bool = False,
+        warnings: bool | Literal["none", "warn", "error"] = False,
+        serialize_as_any: bool = False,
+    ) -> str:
+        """Override model_dump_json to handle custom JSON output."""
+        return super().model_dump_json(
+            indent=indent,
+            include=include,
+            exclude=exclude,
+            context=context,
+            by_alias=by_alias,
+            exclude_unset=exclude_unset,
+            exclude_defaults=exclude_defaults,
+            exclude_none=exclude_none,
+            round_trip=round_trip,
+            warnings=warnings,
+            serialize_as_any=serialize_as_any,
+        )
 
     def to_dict(self) -> Dict[str, Any]:
         """Convert json_output and pydantic_output to a dictionary."""

src/crewai/flow/__init__.py

@@ -1,3 +1,5 @@
-from crewai.flow.flow import Flow
+from crewai.flow.flow import Flow, start, listen, or_, and_, router
+from crewai.flow.persistence import persist
+
+__all__ = ["Flow", "start", "listen", "or_", "and_", "router", "persist"]
 
-__all__ = ["Flow"]

src/crewai/flow/flow.py

@@ -1,5 +1,6 @@
 import asyncio
 import inspect
+import logging
 from typing import (
     Any,
     Callable,
@@ -25,15 +26,70 @@ from crewai.flow.flow_events import (
     MethodExecutionStartedEvent,
 )
 from crewai.flow.flow_visualizer import plot_flow
+from crewai.flow.persistence.base import FlowPersistence
 from crewai.flow.utils import get_possible_return_constants
 from crewai.telemetry import Telemetry
+from crewai.utilities.printer import Printer
+
+logger = logging.getLogger(__name__)
 
 
 class FlowState(BaseModel):
     """Base model for all flow states, ensuring each state has a unique ID."""
-    id: str = Field(default_factory=lambda: str(uuid4()), description="Unique identifier for the flow state")
 
-T = TypeVar("T", bound=Union[FlowState, Dict[str, Any]])
+    id: str = Field(
+        default_factory=lambda: str(uuid4()),
+        description="Unique identifier for the flow state",
+    )
+
+
+# Type variables with explicit bounds
+T = TypeVar(
+    "T", bound=Union[Dict[str, Any], BaseModel]
+)  # Generic flow state type parameter
+StateT = TypeVar(
+    "StateT", bound=Union[Dict[str, Any], BaseModel]
+)  # State validation type parameter
+
+
+def ensure_state_type(state: Any, expected_type: Type[StateT]) -> StateT:
+    """Ensure state matches expected type with proper validation.
+
+    Args:
+        state: State instance to validate
+        expected_type: Expected type for the state
+
+    Returns:
+        Validated state instance
+
+    Raises:
+        TypeError: If state doesn't match expected type
+        ValueError: If state validation fails
+    """
+    """Ensure state matches expected type with proper validation.
+
+    Args:
+        state: State instance to validate
+        expected_type: Expected type for the state
+
+    Returns:
+        Validated state instance
+
+    Raises:
+        TypeError: If state doesn't match expected type
+        ValueError: If state validation fails
+    """
+    if expected_type is dict:
+        if not isinstance(state, dict):
+            raise TypeError(f"Expected dict, got {type(state).__name__}")
+        return cast(StateT, state)
+    if isinstance(expected_type, type) and issubclass(expected_type, BaseModel):
+        if not isinstance(state, expected_type):
+            raise TypeError(
+                f"Expected {expected_type.__name__}, got {type(state).__name__}"
+            )
+        return cast(StateT, state)
+    raise TypeError(f"Invalid expected_type: {expected_type}")
 
 
 def start(condition: Optional[Union[str, dict, Callable]] = None) -> Callable:
@@ -77,6 +133,7 @@ def start(condition: Optional[Union[str, dict, Callable]] = None) -> Callable:
     >>> def complex_start(self):
     ...     pass
     """
+
     def decorator(func):
         func.__is_start_method__ = True
         if condition is not None:
@@ -101,6 +158,7 @@ def start(condition: Optional[Union[str, dict, Callable]] = None) -> Callable:
 
     return decorator
 
+
 def listen(condition: Union[str, dict, Callable]) -> Callable:
     """
     Creates a listener that executes when specified conditions are met.
@@ -137,6 +195,7 @@ def listen(condition: Union[str, dict, Callable]) -> Callable:
     >>> def handle_completion(self):
     ...     pass
     """
+
     def decorator(func):
         if isinstance(condition, str):
             func.__trigger_methods__ = [condition]
@@ -201,6 +260,7 @@ def router(condition: Union[str, dict, Callable]) -> Callable:
     ...         return CONTINUE
     ...     return STOP
     """
+
     def decorator(func):
         func.__is_router__ = True
         if isinstance(condition, str):
@@ -224,6 +284,7 @@ def router(condition: Union[str, dict, Callable]) -> Callable:
 
     return decorator
 
+
 def or_(*conditions: Union[str, dict, Callable]) -> dict:
     """
     Combines multiple conditions with OR logic for flow control.
@@ -326,21 +387,32 @@ class FlowMeta(type):
         routers = set()
 
         for attr_name, attr_value in dct.items():
-            if hasattr(attr_value, "__is_start_method__"):
-                start_methods.append(attr_name)
+            # Check for any flow-related attributes
+            if (
+                hasattr(attr_value, "__is_flow_method__")
+                or hasattr(attr_value, "__is_start_method__")
+                or hasattr(attr_value, "__trigger_methods__")
+                or hasattr(attr_value, "__is_router__")
+            ):
+
+                # Register start methods
+                if hasattr(attr_value, "__is_start_method__"):
+                    start_methods.append(attr_name)
+
+                # Register listeners and routers
                 if hasattr(attr_value, "__trigger_methods__"):
                     methods = attr_value.__trigger_methods__
                     condition_type = getattr(attr_value, "__condition_type__", "OR")
                     listeners[attr_name] = (condition_type, methods)
-            elif hasattr(attr_value, "__trigger_methods__"):
-                methods = attr_value.__trigger_methods__
-                condition_type = getattr(attr_value, "__condition_type__", "OR")
-                listeners[attr_name] = (condition_type, methods)
-                if hasattr(attr_value, "__is_router__") and attr_value.__is_router__:
-                    routers.add(attr_name)
-                    possible_returns = get_possible_return_constants(attr_value)
-                    if possible_returns:
-                        router_paths[attr_name] = possible_returns
+
+                    if (
+                        hasattr(attr_value, "__is_router__")
+                        and attr_value.__is_router__
+                    ):
+                        routers.add(attr_name)
+                        possible_returns = get_possible_return_constants(attr_value)
+                        if possible_returns:
+                            router_paths[attr_name] = possible_returns
 
         setattr(cls, "_start_methods", start_methods)
         setattr(cls, "_listeners", listeners)
@@ -351,7 +423,12 @@ class FlowMeta(type):
 
 
 class Flow(Generic[T], metaclass=FlowMeta):
+    """Base class for all flows.
+
+    Type parameter T must be either Dict[str, Any] or a subclass of BaseModel."""
+
     _telemetry = Telemetry()
+    _printer = Printer()
 
     _start_methods: List[str] = []
     _listeners: Dict[str, tuple[str, List[str]]] = {}
@@ -367,53 +444,130 @@ class Flow(Generic[T], metaclass=FlowMeta):
         _FlowGeneric.__name__ = f"{cls.__name__}[{item.__name__}]"
         return _FlowGeneric
 
-    def __init__(self) -> None:
+    def __init__(
+        self,
+        persistence: Optional[FlowPersistence] = None,
+        **kwargs: Any,
+    ) -> None:
+        """Initialize a new Flow instance.
+
+        Args:
+            persistence: Optional persistence backend for storing flow states
+            **kwargs: Additional state values to initialize or override
+        """
+        # Initialize basic instance attributes
         self._methods: Dict[str, Callable] = {}
-        self._state: T = self._create_initial_state()
         self._method_execution_counts: Dict[str, int] = {}
         self._pending_and_listeners: Dict[str, Set[str]] = {}
         self._method_outputs: List[Any] = []  # List to store all method outputs
+        self._persistence: Optional[FlowPersistence] = persistence
+
+        # Initialize state with initial values
+        self._state = self._create_initial_state()
+
+        # Apply any additional kwargs
+        if kwargs:
+            self._initialize_state(kwargs)
 
         self._telemetry.flow_creation_span(self.__class__.__name__)
 
+        # Register all flow-related methods
         for method_name in dir(self):
-            if callable(getattr(self, method_name)) and not method_name.startswith(
-                "__"
-            ):
-                self._methods[method_name] = getattr(self, method_name)
+            if not method_name.startswith("_"):
+                method = getattr(self, method_name)
+                # Check for any flow-related attributes
+                if (
+                    hasattr(method, "__is_flow_method__")
+                    or hasattr(method, "__is_start_method__")
+                    or hasattr(method, "__trigger_methods__")
+                    or hasattr(method, "__is_router__")
+                ):
+                    # Ensure method is bound to this instance
+                    if not hasattr(method, "__self__"):
+                        method = method.__get__(self, self.__class__)
+                    self._methods[method_name] = method
 
     def _create_initial_state(self) -> T:
+        """Create and initialize flow state with UUID and default values.
+
+        Returns:
+            New state instance with UUID and default values initialized
+
+        Raises:
+            ValueError: If structured state model lacks 'id' field
+            TypeError: If state is neither BaseModel nor dictionary
+        """
         # Handle case where initial_state is None but we have a type parameter
         if self.initial_state is None and hasattr(self, "_initial_state_T"):
             state_type = getattr(self, "_initial_state_T")
             if isinstance(state_type, type):
                 if issubclass(state_type, FlowState):
-                    return state_type()  # type: ignore
+                    # Create instance without id, then set it
+                    instance = state_type()
+                    if not hasattr(instance, "id"):
+                        setattr(instance, "id", str(uuid4()))
+                    return cast(T, instance)
                 elif issubclass(state_type, BaseModel):
                     # Create a new type that includes the ID field
                     class StateWithId(state_type, FlowState):  # type: ignore
                         pass
-                    return StateWithId()  # type: ignore
+
+                    instance = StateWithId()
+                    if not hasattr(instance, "id"):
+                        setattr(instance, "id", str(uuid4()))
+                    return cast(T, instance)
+                elif state_type is dict:
+                    return cast(T, {"id": str(uuid4())})
 
         # Handle case where no initial state is provided
         if self.initial_state is None:
-            return {"id": str(uuid4())}  # type: ignore
+            return cast(T, {"id": str(uuid4())})
 
         # Handle case where initial_state is a type (class)
         if isinstance(self.initial_state, type):
             if issubclass(self.initial_state, FlowState):
-                return self.initial_state()  # type: ignore
+                return cast(T, self.initial_state())  # Uses model defaults
             elif issubclass(self.initial_state, BaseModel):
-                # Create a new type that includes the ID field
-                class StateWithId(self.initial_state, FlowState):  # type: ignore
-                    pass
-                return StateWithId()  # type: ignore
+                # Validate that the model has an id field
+                model_fields = getattr(self.initial_state, "model_fields", None)
+                if not model_fields or "id" not in model_fields:
+                    raise ValueError("Flow state model must have an 'id' field")
+                return cast(T, self.initial_state())  # Uses model defaults
+            elif self.initial_state is dict:
+                return cast(T, {"id": str(uuid4())})
 
-        # Handle dictionary case
-        if isinstance(self.initial_state, dict) and "id" not in self.initial_state:
-            self.initial_state["id"] = str(uuid4())
+        # Handle dictionary instance case
+        if isinstance(self.initial_state, dict):
+            new_state = dict(self.initial_state)  # Copy to avoid mutations
+            if "id" not in new_state:
+                new_state["id"] = str(uuid4())
+            return cast(T, new_state)
 
-        return self.initial_state  # type: ignore
+        # Handle BaseModel instance case
+        if isinstance(self.initial_state, BaseModel):
+            model = cast(BaseModel, self.initial_state)
+            if not hasattr(model, "id"):
+                raise ValueError("Flow state model must have an 'id' field")
+
+            # Create new instance with same values to avoid mutations
+            if hasattr(model, "model_dump"):
+                # Pydantic v2
+                state_dict = model.model_dump()
+            elif hasattr(model, "dict"):
+                # Pydantic v1
+                state_dict = model.dict()
+            else:
+                # Fallback for other BaseModel implementations
+                state_dict = {
+                    k: v for k, v in model.__dict__.items() if not k.startswith("_")
+                }
+
+            # Create new instance of the same class
+            model_class = type(model)
+            return cast(T, model_class(**state_dict))
+        raise TypeError(
+            f"Initial state must be dict or BaseModel, got {type(self.initial_state)}"
+        )
 
     @property
     def state(self) -> T:
@@ -424,53 +578,163 @@ class Flow(Generic[T], metaclass=FlowMeta):
         """Returns the list of all outputs from executed methods."""
         return self._method_outputs
 
+    @property
+    def flow_id(self) -> str:
+        """Returns the unique identifier of this flow instance.
+
+        This property provides a consistent way to access the flow's unique identifier
+        regardless of the underlying state implementation (dict or BaseModel).
+
+        Returns:
+            str: The flow's unique identifier, or an empty string if not found
+
+        Note:
+            This property safely handles both dictionary and BaseModel state types,
+            returning an empty string if the ID cannot be retrieved rather than raising
+            an exception.
+
+        Example:
+            ```python
+            flow = MyFlow()
+            print(f"Current flow ID: {flow.flow_id}")  # Safely get flow ID
+            ```
+        """
+        try:
+            if not hasattr(self, "_state"):
+                return ""
+
+            if isinstance(self._state, dict):
+                return str(self._state.get("id", ""))
+            elif isinstance(self._state, BaseModel):
+                return str(getattr(self._state, "id", ""))
+            return ""
+        except (AttributeError, TypeError):
+            return ""  # Safely handle any unexpected attribute access issues
+
     def _initialize_state(self, inputs: Dict[str, Any]) -> None:
+        """Initialize or update flow state with new inputs.
+
+        Args:
+            inputs: Dictionary of state values to set/update
+
+        Raises:
+            ValueError: If validation fails for structured state
+            TypeError: If state is neither BaseModel nor dictionary
+        """
         if isinstance(self._state, dict):
-            # Preserve the ID when updating unstructured state
+            # For dict states, preserve existing fields unless overridden
             current_id = self._state.get("id")
-            self._state.update(inputs)
+            # Only update specified fields
+            for k, v in inputs.items():
+                self._state[k] = v
+            # Ensure ID is preserved or generated
             if current_id:
                 self._state["id"] = current_id
             elif "id" not in self._state:
                 self._state["id"] = str(uuid4())
         elif isinstance(self._state, BaseModel):
-            # Structured state
+            # For BaseModel states, preserve existing fields unless overridden
             try:
-                def create_model_with_extra_forbid(
-                    base_model: Type[BaseModel],
-                ) -> Type[BaseModel]:
-                    class ModelWithExtraForbid(base_model):  # type: ignore
-                        model_config = base_model.model_config.copy()
-                        model_config["extra"] = "forbid"
-
-                    return ModelWithExtraForbid
-
-                # Get current state as dict, preserving the ID if it exists
-                state_model = cast(BaseModel, self._state)
-                current_state = (
-                    state_model.model_dump()
-                    if hasattr(state_model, "model_dump")
-                    else state_model.dict()
-                    if hasattr(state_model, "dict")
-                    else {
-                        k: v
-                        for k, v in state_model.__dict__.items()
-                        if not k.startswith("_")
+                model = cast(BaseModel, self._state)
+                # Get current state as dict
+                if hasattr(model, "model_dump"):
+                    current_state = model.model_dump()
+                elif hasattr(model, "dict"):
+                    current_state = model.dict()
+                else:
+                    current_state = {
+                        k: v for k, v in model.__dict__.items() if not k.startswith("_")
                     }
-                )
 
-                ModelWithExtraForbid = create_model_with_extra_forbid(
-                    self._state.__class__
-                )
-                self._state = cast(
-                    T, ModelWithExtraForbid(**{**current_state, **inputs})
-                )
+                # Create new state with preserved fields and updates
+                new_state = {**current_state, **inputs}
+
+                # Create new instance with merged state
+                model_class = type(model)
+                if hasattr(model_class, "model_validate"):
+                    # Pydantic v2
+                    self._state = cast(T, model_class.model_validate(new_state))
+                elif hasattr(model_class, "parse_obj"):
+                    # Pydantic v1
+                    self._state = cast(T, model_class.parse_obj(new_state))
+                else:
+                    # Fallback for other BaseModel implementations
+                    self._state = cast(T, model_class(**new_state))
             except ValidationError as e:
                 raise ValueError(f"Invalid inputs for structured state: {e}") from e
         else:
             raise TypeError("State must be a BaseModel instance or a dictionary.")
 
+    def _restore_state(self, stored_state: Dict[str, Any]) -> None:
+        """Restore flow state from persistence.
+
+        Args:
+            stored_state: Previously stored state to restore
+
+        Raises:
+            ValueError: If validation fails for structured state
+            TypeError: If state is neither BaseModel nor dictionary
+        """
+        # When restoring from persistence, use the stored ID
+        stored_id = stored_state.get("id")
+        if not stored_id:
+            raise ValueError("Stored state must have an 'id' field")
+
+        if isinstance(self._state, dict):
+            # For dict states, update all fields from stored state
+            self._state.clear()
+            self._state.update(stored_state)
+        elif isinstance(self._state, BaseModel):
+            # For BaseModel states, create new instance with stored values
+            model = cast(BaseModel, self._state)
+            if hasattr(model, "model_validate"):
+                # Pydantic v2
+                self._state = cast(T, type(model).model_validate(stored_state))
+            elif hasattr(model, "parse_obj"):
+                # Pydantic v1
+                self._state = cast(T, type(model).parse_obj(stored_state))
+            else:
+                # Fallback for other BaseModel implementations
+                self._state = cast(T, type(model)(**stored_state))
+        else:
+            raise TypeError(f"State must be dict or BaseModel, got {type(self._state)}")
+
     def kickoff(self, inputs: Optional[Dict[str, Any]] = None) -> Any:
+        """Start the flow execution.
+
+        Args:
+            inputs: Optional dictionary containing input values and potentially a state ID to restore
+        """
+        # Handle state restoration if ID is provided in inputs
+        if inputs and "id" in inputs and self._persistence is not None:
+            restore_uuid = inputs["id"]
+            stored_state = self._persistence.load_state(restore_uuid)
+
+            # Override the id in the state if it exists in inputs
+            if "id" in inputs:
+                if isinstance(self._state, dict):
+                    self._state["id"] = inputs["id"]
+                elif isinstance(self._state, BaseModel):
+                    setattr(self._state, "id", inputs["id"])
+
+            if stored_state:
+                self._log_flow_event(
+                    f"Loading flow state from memory for UUID: {restore_uuid}",
+                    color="yellow",
+                )
+                # Restore the state
+                self._restore_state(stored_state)
+            else:
+                self._log_flow_event(
+                    f"No flow state found for UUID: {restore_uuid}", color="red"
+                )
+
+            # Apply any additional inputs after restoration
+            filtered_inputs = {k: v for k, v in inputs.items() if k != "id"}
+            if filtered_inputs:
+                self._initialize_state(filtered_inputs)
+
+        # Start flow execution
         self.event_emitter.send(
             self,
             event=FlowStartedEvent(
@@ -478,9 +742,13 @@ class Flow(Generic[T], metaclass=FlowMeta):
                 flow_name=self.__class__.__name__,
             ),
         )
+        self._log_flow_event(
+            f"Flow started with ID: {self.flow_id}", color="bold_magenta"
+        )
 
-        if inputs is not None:
+        if inputs is not None and "id" not in inputs:
             self._initialize_state(inputs)
+
         return asyncio.run(self.kickoff_async())
 
     async def kickoff_async(self, inputs: Optional[Dict[str, Any]] = None) -> Any:
@@ -723,6 +991,32 @@ class Flow(Generic[T], metaclass=FlowMeta):
 
             traceback.print_exc()
 
+    def _log_flow_event(
+        self, message: str, color: str = "yellow", level: str = "info"
+    ) -> None:
+        """Centralized logging method for flow events.
+
+        This method provides a consistent interface for logging flow-related events,
+        combining both console output with colors and proper logging levels.
+
+        Args:
+            message: The message to log
+            color: Color to use for console output (default: yellow)
+                  Available colors: purple, red, bold_green, bold_purple,
+                  bold_blue, yellow, yellow
+            level: Log level to use (default: info)
+                  Supported levels: info, warning
+
+        Note:
+            This method uses the Printer utility for colored console output
+            and the standard logging module for log level support.
+        """
+        self._printer.print(message, color=color)
+        if level == "info":
+            logger.info(message)
+        elif level == "warning":
+            logger.warning(message)
+
     def plot(self, filename: str = "crewai_flow") -> None:
         self._telemetry.flow_plotting_span(
             self.__class__.__name__, list(self._methods.keys())

src/crewai/flow/persistence/__init__.py

@@ -0,0 +1,18 @@
+"""
+CrewAI Flow Persistence.
+
+This module provides interfaces and implementations for persisting flow states.
+"""
+
+from typing import Any, Dict, TypeVar, Union
+
+from pydantic import BaseModel
+
+from crewai.flow.persistence.base import FlowPersistence
+from crewai.flow.persistence.decorators import persist
+from crewai.flow.persistence.sqlite import SQLiteFlowPersistence
+
+__all__ = ["FlowPersistence", "persist", "SQLiteFlowPersistence"]
+
+StateType = TypeVar('StateType', bound=Union[Dict[str, Any], BaseModel])
+DictStateType = Dict[str, Any]

src/crewai/flow/persistence/base.py

@@ -0,0 +1,53 @@
+"""Base class for flow state persistence."""
+
+import abc
+from typing import Any, Dict, Optional, Union
+
+from pydantic import BaseModel
+
+
+class FlowPersistence(abc.ABC):
+    """Abstract base class for flow state persistence.
+    
+    This class defines the interface that all persistence implementations must follow.
+    It supports both structured (Pydantic BaseModel) and unstructured (dict) states.
+    """
+    
+    @abc.abstractmethod
+    def init_db(self) -> None:
+        """Initialize the persistence backend.
+        
+        This method should handle any necessary setup, such as:
+        - Creating tables
+        - Establishing connections
+        - Setting up indexes
+        """
+        pass
+    
+    @abc.abstractmethod
+    def save_state(
+        self,
+        flow_uuid: str,
+        method_name: str,
+        state_data: Union[Dict[str, Any], BaseModel]
+    ) -> None:
+        """Persist the flow state after method completion.
+        
+        Args:
+            flow_uuid: Unique identifier for the flow instance
+            method_name: Name of the method that just completed
+            state_data: Current state data (either dict or Pydantic model)
+        """
+        pass
+    
+    @abc.abstractmethod
+    def load_state(self, flow_uuid: str) -> Optional[Dict[str, Any]]:
+        """Load the most recent state for a given flow UUID.
+        
+        Args:
+            flow_uuid: Unique identifier for the flow instance
+            
+        Returns:
+            The most recent state as a dictionary, or None if no state exists
+        """
+        pass

src/crewai/flow/persistence/decorators.py

@@ -0,0 +1,252 @@
+"""
+Decorators for flow state persistence.
+
+Example:
+    ```python
+    from crewai.flow.flow import Flow, start
+    from crewai.flow.persistence import persist, SQLiteFlowPersistence
+
+    class MyFlow(Flow):
+        @start()
+        @persist(SQLiteFlowPersistence())
+        def sync_method(self):
+            # Synchronous method implementation
+            pass
+
+        @start()
+        @persist(SQLiteFlowPersistence())
+        async def async_method(self):
+            # Asynchronous method implementation
+            await some_async_operation()
+    ```
+"""
+
+import asyncio
+import functools
+import logging
+from typing import (
+    Any,
+    Callable,
+    Optional,
+    Type,
+    TypeVar,
+    Union,
+    cast,
+)
+
+from pydantic import BaseModel
+
+from crewai.flow.persistence.base import FlowPersistence
+from crewai.flow.persistence.sqlite import SQLiteFlowPersistence
+from crewai.utilities.printer import Printer
+
+logger = logging.getLogger(__name__)
+T = TypeVar("T")
+
+# Constants for log messages
+LOG_MESSAGES = {
+    "save_state": "Saving flow state to memory for ID: {}",
+    "save_error": "Failed to persist state for method {}: {}",
+    "state_missing": "Flow instance has no state",
+    "id_missing": "Flow state must have an 'id' field for persistence"
+}
+
+
+class PersistenceDecorator:
+    """Class to handle flow state persistence with consistent logging."""
+
+    _printer = Printer()  # Class-level printer instance
+
+    @classmethod
+    def persist_state(cls, flow_instance: Any, method_name: str, persistence_instance: FlowPersistence) -> None:
+        """Persist flow state with proper error handling and logging.
+
+        This method handles the persistence of flow state data, including proper
+        error handling and colored console output for status updates.
+
+        Args:
+            flow_instance: The flow instance whose state to persist
+            method_name: Name of the method that triggered persistence
+            persistence_instance: The persistence backend to use
+
+        Raises:
+            ValueError: If flow has no state or state lacks an ID
+            RuntimeError: If state persistence fails
+            AttributeError: If flow instance lacks required state attributes
+        """
+        try:
+            state = getattr(flow_instance, 'state', None)
+            if state is None:
+                raise ValueError("Flow instance has no state")
+
+            flow_uuid: Optional[str] = None
+            if isinstance(state, dict):
+                flow_uuid = state.get('id')
+            elif isinstance(state, BaseModel):
+                flow_uuid = getattr(state, 'id', None)
+
+            if not flow_uuid:
+                raise ValueError("Flow state must have an 'id' field for persistence")
+
+            # Log state saving with consistent message
+            cls._printer.print(LOG_MESSAGES["save_state"].format(flow_uuid), color="cyan")
+            logger.info(LOG_MESSAGES["save_state"].format(flow_uuid))
+
+            try:
+                persistence_instance.save_state(
+                    flow_uuid=flow_uuid,
+                    method_name=method_name,
+                    state_data=state,
+                )
+            except Exception as e:
+                error_msg = LOG_MESSAGES["save_error"].format(method_name, str(e))
+                cls._printer.print(error_msg, color="red")
+                logger.error(error_msg)
+                raise RuntimeError(f"State persistence failed: {str(e)}") from e
+        except AttributeError:
+            error_msg = LOG_MESSAGES["state_missing"]
+            cls._printer.print(error_msg, color="red")
+            logger.error(error_msg)
+            raise ValueError(error_msg)
+        except (TypeError, ValueError) as e:
+            error_msg = LOG_MESSAGES["id_missing"]
+            cls._printer.print(error_msg, color="red")
+            logger.error(error_msg)
+            raise ValueError(error_msg) from e
+
+
+def persist(persistence: Optional[FlowPersistence] = None):
+    """Decorator to persist flow state.
+
+    This decorator can be applied at either the class level or method level.
+    When applied at the class level, it automatically persists all flow method
+    states. When applied at the method level, it persists only that method's
+    state.
+
+    Args:
+        persistence: Optional FlowPersistence implementation to use.
+                    If not provided, uses SQLiteFlowPersistence.
+
+    Returns:
+        A decorator that can be applied to either a class or method
+
+    Raises:
+        ValueError: If the flow state doesn't have an 'id' field
+        RuntimeError: If state persistence fails
+
+    Example:
+        @persist  # Class-level persistence with default SQLite
+        class MyFlow(Flow[MyState]):
+            @start()
+            def begin(self):
+                pass
+    """
+
+    def decorator(target: Union[Type, Callable[..., T]]) -> Union[Type, Callable[..., T]]:
+        """Decorator that handles both class and method decoration."""
+        actual_persistence = persistence or SQLiteFlowPersistence()
+
+        if isinstance(target, type):
+            # Class decoration
+            original_init = getattr(target, "__init__")
+
+            @functools.wraps(original_init)
+            def new_init(self: Any, *args: Any, **kwargs: Any) -> None:
+                if 'persistence' not in kwargs:
+                    kwargs['persistence'] = actual_persistence
+                original_init(self, *args, **kwargs)
+
+            setattr(target, "__init__", new_init)
+
+            # Store original methods to preserve their decorators
+            original_methods = {}
+
+            for name, method in target.__dict__.items():
+                if callable(method) and (
+                    hasattr(method, "__is_start_method__") or
+                    hasattr(method, "__trigger_methods__") or
+                    hasattr(method, "__condition_type__") or
+                    hasattr(method, "__is_flow_method__") or
+                    hasattr(method, "__is_router__")
+                ):
+                    original_methods[name] = method
+
+            # Create wrapped versions of the methods that include persistence
+            for name, method in original_methods.items():
+                if asyncio.iscoroutinefunction(method):
+                    # Create a closure to capture the current name and method
+                    def create_async_wrapper(method_name: str, original_method: Callable):
+                        @functools.wraps(original_method)
+                        async def method_wrapper(self: Any, *args: Any, **kwargs: Any) -> Any:
+                            result = await original_method(self, *args, **kwargs)
+                            PersistenceDecorator.persist_state(self, method_name, actual_persistence)
+                            return result
+                        return method_wrapper
+
+                    wrapped = create_async_wrapper(name, method)
+
+                    # Preserve all original decorators and attributes
+                    for attr in ["__is_start_method__", "__trigger_methods__", "__condition_type__", "__is_router__"]:
+                        if hasattr(method, attr):
+                            setattr(wrapped, attr, getattr(method, attr))
+                    setattr(wrapped, "__is_flow_method__", True)
+
+                    # Update the class with the wrapped method
+                    setattr(target, name, wrapped)
+                else:
+                    # Create a closure to capture the current name and method
+                    def create_sync_wrapper(method_name: str, original_method: Callable):
+                        @functools.wraps(original_method)
+                        def method_wrapper(self: Any, *args: Any, **kwargs: Any) -> Any:
+                            result = original_method(self, *args, **kwargs)
+                            PersistenceDecorator.persist_state(self, method_name, actual_persistence)
+                            return result
+                        return method_wrapper
+
+                    wrapped = create_sync_wrapper(name, method)
+
+                    # Preserve all original decorators and attributes
+                    for attr in ["__is_start_method__", "__trigger_methods__", "__condition_type__", "__is_router__"]:
+                        if hasattr(method, attr):
+                            setattr(wrapped, attr, getattr(method, attr))
+                    setattr(wrapped, "__is_flow_method__", True)
+
+                    # Update the class with the wrapped method
+                    setattr(target, name, wrapped)
+
+            return target
+        else:
+            # Method decoration
+            method = target
+            setattr(method, "__is_flow_method__", True)
+
+            if asyncio.iscoroutinefunction(method):
+                @functools.wraps(method)
+                async def method_async_wrapper(flow_instance: Any, *args: Any, **kwargs: Any) -> T:
+                    method_coro = method(flow_instance, *args, **kwargs)
+                    if asyncio.iscoroutine(method_coro):
+                        result = await method_coro
+                    else:
+                        result = method_coro
+                    PersistenceDecorator.persist_state(flow_instance, method.__name__, actual_persistence)
+                    return result
+
+                for attr in ["__is_start_method__", "__trigger_methods__", "__condition_type__", "__is_router__"]:
+                    if hasattr(method, attr):
+                        setattr(method_async_wrapper, attr, getattr(method, attr))
+                setattr(method_async_wrapper, "__is_flow_method__", True)
+                return cast(Callable[..., T], method_async_wrapper)
+            else:
+                @functools.wraps(method)
+                def method_sync_wrapper(flow_instance: Any, *args: Any, **kwargs: Any) -> T:
+                    result = method(flow_instance, *args, **kwargs)
+                    PersistenceDecorator.persist_state(flow_instance, method.__name__, actual_persistence)
+                    return result
+
+                for attr in ["__is_start_method__", "__trigger_methods__", "__condition_type__", "__is_router__"]:
+                    if hasattr(method, attr):
+                        setattr(method_sync_wrapper, attr, getattr(method, attr))
+                setattr(method_sync_wrapper, "__is_flow_method__", True)
+                return cast(Callable[..., T], method_sync_wrapper)
+
+    return decorator

src/crewai/flow/persistence/sqlite.py

@@ -0,0 +1,123 @@
+"""
+SQLite-based implementation of flow state persistence.
+"""
+
+import json
+import sqlite3
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, Optional, Union
+
+from pydantic import BaseModel
+
+from crewai.flow.persistence.base import FlowPersistence
+
+
+class SQLiteFlowPersistence(FlowPersistence):
+    """SQLite-based implementation of flow state persistence.
+
+    This class provides a simple, file-based persistence implementation using SQLite.
+    It's suitable for development and testing, or for production use cases with
+    moderate performance requirements.
+    """
+
+    db_path: str  # Type annotation for instance variable
+
+    def __init__(self, db_path: Optional[str] = None):
+        """Initialize SQLite persistence.
+
+        Args:
+            db_path: Path to the SQLite database file. If not provided, uses
+                    db_storage_path() from utilities.paths.
+
+        Raises:
+            ValueError: If db_path is invalid
+        """
+        from crewai.utilities.paths import db_storage_path
+        # Get path from argument or default location
+        path = db_path or str(Path(db_storage_path()) / "flow_states.db")
+
+        if not path:
+            raise ValueError("Database path must be provided")
+
+        self.db_path = path  # Now mypy knows this is str
+        self.init_db()
+
+    def init_db(self) -> None:
+        """Create the necessary tables if they don't exist."""
+        with sqlite3.connect(self.db_path) as conn:
+            conn.execute("""
+            CREATE TABLE IF NOT EXISTS flow_states (
+                id INTEGER PRIMARY KEY AUTOINCREMENT,
+                flow_uuid TEXT NOT NULL,
+                method_name TEXT NOT NULL,
+                timestamp DATETIME NOT NULL,
+                state_json TEXT NOT NULL
+            )
+            """)
+            # Add index for faster UUID lookups
+            conn.execute("""
+            CREATE INDEX IF NOT EXISTS idx_flow_states_uuid
+            ON flow_states(flow_uuid)
+            """)
+
+    def save_state(
+        self,
+        flow_uuid: str,
+        method_name: str,
+        state_data: Union[Dict[str, Any], BaseModel],
+    ) -> None:
+        """Save the current flow state to SQLite.
+
+        Args:
+            flow_uuid: Unique identifier for the flow instance
+            method_name: Name of the method that just completed
+            state_data: Current state data (either dict or Pydantic model)
+        """
+        # Convert state_data to dict, handling both Pydantic and dict cases
+        if isinstance(state_data, BaseModel):
+            state_dict = dict(state_data)  # Use dict() for better type compatibility
+        elif isinstance(state_data, dict):
+            state_dict = state_data
+        else:
+            raise ValueError(
+                f"state_data must be either a Pydantic BaseModel or dict, got {type(state_data)}"
+            )
+
+        with sqlite3.connect(self.db_path) as conn:
+            conn.execute("""
+            INSERT INTO flow_states (
+                flow_uuid,
+                method_name,
+                timestamp,
+                state_json
+            ) VALUES (?, ?, ?, ?)
+            """, (
+                flow_uuid,
+                method_name,
+                datetime.utcnow().isoformat(),
+                json.dumps(state_dict),
+            ))
+
+    def load_state(self, flow_uuid: str) -> Optional[Dict[str, Any]]:
+        """Load the most recent state for a given flow UUID.
+
+        Args:
+            flow_uuid: Unique identifier for the flow instance
+
+        Returns:
+            The most recent state as a dictionary, or None if no state exists
+        """
+        with sqlite3.connect(self.db_path) as conn:
+            cursor = conn.execute("""
+            SELECT state_json
+            FROM flow_states
+            WHERE flow_uuid = ?
+            ORDER BY id DESC
+            LIMIT 1
+            """, (flow_uuid,))
+            row = cursor.fetchone()
+
+        if row:
+            return json.loads(row[0])
+        return None

src/crewai/knowledge/embedder/fastembed.py

@@ -47,7 +47,7 @@ class FastEmbed(BaseEmbedder):
             cache_dir=str(cache_dir) if cache_dir else None,
         )
 
-    def embed_chunks(self, chunks: List[str]) -> List[np.ndarray]:
+    def embed_chunks(self, chunks: List[str]) -> np.ndarray:
         """
         Generate embeddings for a list of text chunks
 
@@ -55,12 +55,12 @@ class FastEmbed(BaseEmbedder):
             chunks: List of text chunks to embed
 
         Returns:
-            List of embeddings
+            Array of embeddings
         """
         embeddings = list(self.model.embed(chunks))
-        return embeddings
+        return np.stack(embeddings)
 
-    def embed_texts(self, texts: List[str]) -> List[np.ndarray]:
+    def embed_texts(self, texts: List[str]) -> np.ndarray:
         """
         Generate embeddings for a list of texts
 
@@ -68,10 +68,10 @@ class FastEmbed(BaseEmbedder):
             texts: List of texts to embed
 
         Returns:
-            List of embeddings
+            Array of embeddings
         """
         embeddings = list(self.model.embed(texts))
-        return embeddings
+        return np.stack(embeddings)
 
     def embed_text(self, text: str) -> np.ndarray:
         """

src/crewai/knowledge/knowledge.py

@@ -15,20 +15,20 @@ class Knowledge(BaseModel):
     Args:
         sources: List[BaseKnowledgeSource] = Field(default_factory=list)
         storage: Optional[KnowledgeStorage] = Field(default=None)
-        embedder_config: Optional[Dict[str, Any]] = None
+        embedder: Optional[Dict[str, Any]] = None
     """
 
     sources: List[BaseKnowledgeSource] = Field(default_factory=list)
     model_config = ConfigDict(arbitrary_types_allowed=True)
     storage: Optional[KnowledgeStorage] = Field(default=None)
-    embedder_config: Optional[Dict[str, Any]] = None
+    embedder: Optional[Dict[str, Any]] = None
     collection_name: Optional[str] = None
 
     def __init__(
         self,
         collection_name: str,
         sources: List[BaseKnowledgeSource],
-        embedder_config: Optional[Dict[str, Any]] = None,
+        embedder: Optional[Dict[str, Any]] = None,
         storage: Optional[KnowledgeStorage] = None,
         **data,
     ):
@@ -37,25 +37,23 @@ class Knowledge(BaseModel):
             self.storage = storage
         else:
             self.storage = KnowledgeStorage(
-                embedder_config=embedder_config, collection_name=collection_name
+                embedder=embedder, collection_name=collection_name
             )
         self.sources = sources
         self.storage.initialize_knowledge_storage()
-        for source in sources:
-            source.storage = self.storage
-            source.add()
+        self._add_sources()
 
     def query(self, query: List[str], limit: int = 3) -> List[Dict[str, Any]]:
         """
         Query across all knowledge sources to find the most relevant information.
         Returns the top_k most relevant chunks.
-        
+
         Raises:
             ValueError: If storage is not initialized.
         """
         if self.storage is None:
             raise ValueError("Storage is not initialized.")
-            
+
         results = self.storage.search(
             query,
             limit,
@@ -63,6 +61,15 @@ class Knowledge(BaseModel):
         return results
 
     def _add_sources(self):
-        for source in self.sources:
-            source.storage = self.storage
-            source.add()
+        try:
+            for source in self.sources:
+                source.storage = self.storage
+                source.add()
+        except Exception as e:
+            raise e
+
+    def reset(self) -> None:
+        if self.storage:
+            self.storage.reset()
+        else:
+            raise ValueError("Storage is not initialized.")

src/crewai/knowledge/source/base_file_knowledge_source.py

@@ -29,7 +29,13 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
     def validate_file_path(cls, v, info):
         """Validate that at least one of file_path or file_paths is provided."""
         # Single check if both are None, O(1) instead of nested conditions
-        if v is None and info.data.get("file_path" if info.field_name == "file_paths" else "file_paths") is None:
+        if (
+            v is None
+            and info.data.get(
+                "file_path" if info.field_name == "file_paths" else "file_paths"
+            )
+            is None
+        ):
             raise ValueError("Either file_path or file_paths must be provided")
         return v
 

src/crewai/knowledge/source/crew_docling_source.py

@@ -8,6 +8,7 @@ try:
     from docling.exceptions import ConversionError
     from docling_core.transforms.chunker.hierarchical_chunker import HierarchicalChunker
     from docling_core.types.doc.document import DoclingDocument
+
     DOCLING_AVAILABLE = True
 except ImportError:
     DOCLING_AVAILABLE = False
@@ -38,8 +39,8 @@ class CrewDoclingSource(BaseKnowledgeSource):
     file_paths: List[Union[Path, str]] = Field(default_factory=list)
     chunks: List[str] = Field(default_factory=list)
     safe_file_paths: List[Union[Path, str]] = Field(default_factory=list)
-    content: List[DoclingDocument] = Field(default_factory=list)
-    document_converter: DocumentConverter = Field(
+    content: List["DoclingDocument"] = Field(default_factory=list)
+    document_converter: "DocumentConverter" = Field(
         default_factory=lambda: DocumentConverter(
             allowed_formats=[
                 InputFormat.MD,
@@ -65,7 +66,7 @@ class CrewDoclingSource(BaseKnowledgeSource):
         self.safe_file_paths = self.validate_content()
         self.content = self._load_content()
 
-    def _load_content(self) -> List[DoclingDocument]:
+    def _load_content(self) -> List["DoclingDocument"]:
         try:
             return self._convert_source_to_docling_documents()
         except ConversionError as e:
@@ -87,11 +88,11 @@ class CrewDoclingSource(BaseKnowledgeSource):
             self.chunks.extend(list(new_chunks_iterable))
         self._save_documents()
 
-    def _convert_source_to_docling_documents(self) -> List[DoclingDocument]:
+    def _convert_source_to_docling_documents(self) -> List["DoclingDocument"]:
         conv_results_iter = self.document_converter.convert_all(self.safe_file_paths)
         return [result.document for result in conv_results_iter]
 
-    def _chunk_doc(self, doc: DoclingDocument) -> Iterator[str]:
+    def _chunk_doc(self, doc: "DoclingDocument") -> Iterator[str]:
         chunker = HierarchicalChunker()
         for chunk in chunker.chunk(doc):
             yield chunk.text

src/crewai/knowledge/storage/knowledge_storage.py

@@ -48,11 +48,11 @@ class KnowledgeStorage(BaseKnowledgeStorage):
 
     def __init__(
         self,
-        embedder_config: Optional[Dict[str, Any]] = None,
+        embedder: Optional[Dict[str, Any]] = None,
         collection_name: Optional[str] = None,
     ):
         self.collection_name = collection_name
-        self._set_embedder_config(embedder_config)
+        self._set_embedder_config(embedder)
 
     def search(
         self,
@@ -99,7 +99,7 @@ class KnowledgeStorage(BaseKnowledgeStorage):
             )
             if self.app:
                 self.collection = self.app.get_or_create_collection(
-                    name=collection_name, embedding_function=self.embedder_config
+                    name=collection_name, embedding_function=self.embedder
                 )
             else:
                 raise Exception("Vector Database Client not initialized")
@@ -154,9 +154,15 @@ class KnowledgeStorage(BaseKnowledgeStorage):
                 filtered_ids.append(doc_id)
 
             # If we have no metadata at all, set it to None
-            final_metadata: Optional[OneOrMany[chromadb.Metadata]] = (
-                None if all(m is None for m in filtered_metadata) else filtered_metadata
-            )
+            final_metadata: Optional[List[Dict[str, Union[str, int, float, bool]]]] = None
+            if not all(m is None for m in filtered_metadata):
+                final_metadata = []
+                for m in filtered_metadata:
+                    if m is not None:
+                        filtered_m = {k: v for k, v in m.items() if isinstance(v, (str, int, float, bool))}
+                        final_metadata.append(filtered_m)
+                    else:
+                        final_metadata.append({"empty": True})
 
             self.collection.upsert(
                 documents=filtered_docs,
@@ -187,17 +193,15 @@ class KnowledgeStorage(BaseKnowledgeStorage):
             api_key=os.getenv("OPENAI_API_KEY"), model_name="text-embedding-3-small"
         )
 
-    def _set_embedder_config(
-        self, embedder_config: Optional[Dict[str, Any]] = None
-    ) -> None:
+    def _set_embedder_config(self, embedder: Optional[Dict[str, Any]] = None) -> None:
         """Set the embedding configuration for the knowledge storage.
 
         Args:
             embedder_config (Optional[Dict[str, Any]]): Configuration dictionary for the embedder.
                 If None or empty, defaults to the default embedding function.
         """
-        self.embedder_config = (
-            EmbeddingConfigurator().configure_embedder(embedder_config)
-            if embedder_config
+        self.embedder = (
+            EmbeddingConfigurator().configure_embedder(embedder)
+            if embedder
             else self._create_default_embedding_function()
         )

src/crewai/llm.py

@@ -5,15 +5,17 @@ import sys
 import threading
 import warnings
 from contextlib import contextmanager
-from typing import Any, Dict, List, Optional, Union, cast
+from typing import Any, Dict, List, Literal, Optional, Type, Union, cast
 
 from dotenv import load_dotenv
+from pydantic import BaseModel
 
 with warnings.catch_warnings():
     warnings.simplefilter("ignore", UserWarning)
     import litellm
     from litellm import Choices, get_supported_openai_params
     from litellm.types.utils import ModelResponse
+    from litellm.utils import supports_response_schema
 
 
 from crewai.utilities.exceptions.context_window_exceeding_exception import (
@@ -128,21 +130,23 @@ class LLM:
         presence_penalty: Optional[float] = None,
         frequency_penalty: Optional[float] = None,
         logit_bias: Optional[Dict[int, float]] = None,
-        response_format: Optional[Dict[str, Any]] = None,
+        response_format: Optional[Type[BaseModel]] = None,
         seed: Optional[int] = None,
         logprobs: Optional[int] = None,
         top_logprobs: Optional[int] = None,
         base_url: Optional[str] = None,
+        api_base: Optional[str] = None,
         api_version: Optional[str] = None,
         api_key: Optional[str] = None,
         callbacks: List[Any] = [],
+        reasoning_effort: Optional[Literal["none", "low", "medium", "high"]] = None,
+        **kwargs,
     ):
         self.model = model
         self.timeout = timeout
         self.temperature = temperature
         self.top_p = top_p
         self.n = n
-        self.stop = stop
         self.max_completion_tokens = max_completion_tokens
         self.max_tokens = max_tokens
         self.presence_penalty = presence_penalty
@@ -153,47 +157,110 @@ class LLM:
         self.logprobs = logprobs
         self.top_logprobs = top_logprobs
         self.base_url = base_url
+        self.api_base = api_base
         self.api_version = api_version
         self.api_key = api_key
         self.callbacks = callbacks
         self.context_window_size = 0
+        self.reasoning_effort = reasoning_effort
+        self.additional_params = kwargs
+        self.is_anthropic = self._is_anthropic_model(model)
 
         litellm.drop_params = True
 
+        # Normalize self.stop to always be a List[str]
+        if stop is None:
+            self.stop: List[str] = []
+        elif isinstance(stop, str):
+            self.stop = [stop]
+        else:
+            self.stop = stop
+
         self.set_callbacks(callbacks)
         self.set_env_callbacks()
 
+    def _is_anthropic_model(self, model: str) -> bool:
+        """Determine if the model is from Anthropic provider.
+        
+        Args:
+            model: The model identifier string.
+            
+        Returns:
+            bool: True if the model is from Anthropic, False otherwise.
+        """
+        ANTHROPIC_PREFIXES = ('anthropic/', 'claude-', 'claude/')
+        return any(prefix in model.lower() for prefix in ANTHROPIC_PREFIXES)
+
     def call(
         self,
-        messages: List[Dict[str, str]],
+        messages: Union[str, List[Dict[str, str]]],
         tools: Optional[List[dict]] = None,
         callbacks: Optional[List[Any]] = None,
         available_functions: Optional[Dict[str, Any]] = None,
-    ) -> str:
+    ) -> Union[str, Any]:
+        """High-level LLM call method.
+        
+        Args:
+            messages: Input messages for the LLM.
+                     Can be a string or list of message dictionaries.
+                     If string, it will be converted to a single user message.
+                     If list, each dict must have 'role' and 'content' keys.
+            tools: Optional list of tool schemas for function calling.
+                  Each tool should define its name, description, and parameters.
+            callbacks: Optional list of callback functions to be executed
+                      during and after the LLM call.
+            available_functions: Optional dict mapping function names to callables
+                               that can be invoked by the LLM.
+        
+        Returns:
+            Union[str, Any]: Either a text response from the LLM (str) or
+                           the result of a tool function call (Any).
+        
+        Raises:
+            TypeError: If messages format is invalid
+            ValueError: If response format is not supported
+            LLMContextLengthExceededException: If input exceeds model's context limit
+        
+        Examples:
+            # Example 1: Simple string input
+            >>> response = llm.call("Return the name of a random city.")
+            >>> print(response)
+            "Paris"
+            
+            # Example 2: Message list with system and user messages
+            >>> messages = [
+            ...     {"role": "system", "content": "You are a geography expert"},
+            ...     {"role": "user", "content": "What is France's capital?"}
+            ... ]
+            >>> response = llm.call(messages)
+            >>> print(response)
+            "The capital of France is Paris."
         """
-        High-level call method that:
-          1) Calls litellm.completion
-          2) Checks for function/tool calls
-          3) If a tool call is found:
-               a) executes the function
-               b) returns the result
-          4) If no tool call, returns the text response
+        # Validate parameters before proceeding with the call.
+        self._validate_call_params()
+
+        if isinstance(messages, str):
+            messages = [{"role": "user", "content": messages}]
+
+        # For O1 models, system messages are not supported.
+        # Convert any system messages into assistant messages.
+        if "o1" in self.model.lower():
+            for message in messages:
+                if message.get("role") == "system":
+                    message["role"] = "assistant"
 
-        :param messages: The conversation messages
-        :param tools: Optional list of function schemas for function calling
-        :param callbacks: Optional list of callbacks
-        :param available_functions: A dictionary mapping function_name -> actual Python function
-        :return: Final text response from the LLM or the tool result
-        """
         with suppress_warnings():
             if callbacks and len(callbacks) > 0:
                 self.set_callbacks(callbacks)
 
             try:
-                # --- 1) Make the completion call
+                # --- 1) Format messages according to provider requirements
+                formatted_messages = self._format_messages_for_provider(messages)
+
+                # --- 2) Prepare the parameters for the completion call
                 params = {
                     "model": self.model,
-                    "messages": messages,
+                    "messages": formatted_messages,
                     "timeout": self.timeout,
                     "temperature": self.temperature,
                     "top_p": self.top_p,
@@ -207,15 +274,20 @@ class LLM:
                     "seed": self.seed,
                     "logprobs": self.logprobs,
                     "top_logprobs": self.top_logprobs,
-                    "api_base": self.base_url,
+                    "api_base": self.api_base,
+                    "base_url": self.base_url,
                     "api_version": self.api_version,
                     "api_key": self.api_key,
                     "stream": False,
-                    "tools": tools,  # pass the tool schema
+                    "tools": tools,
+                    "reasoning_effort": self.reasoning_effort,
+                    **self.additional_params,
                 }
 
+                # Remove None values from params
                 params = {k: v for k, v in params.items() if v is not None}
 
+                # --- 2) Make the completion call
                 response = litellm.completion(**params)
                 response_message = cast(Choices, cast(ModelResponse, response).choices)[
                     0
@@ -223,11 +295,24 @@ class LLM:
                 text_response = response_message.content or ""
                 tool_calls = getattr(response_message, "tool_calls", [])
 
-                # --- 2) If no tool calls, return the text response
+                # --- 3) Handle callbacks with usage info
+                if callbacks and len(callbacks) > 0:
+                    for callback in callbacks:
+                        if hasattr(callback, "log_success_event"):
+                            usage_info = getattr(response, "usage", None)
+                            if usage_info:
+                                callback.log_success_event(
+                                    kwargs=params,
+                                    response_obj={"usage": usage_info},
+                                    start_time=0,
+                                    end_time=0,
+                                )
+
+                # --- 4) If no tool calls, return the text response
                 if not tool_calls or not available_functions:
                     return text_response
 
-                # --- 3) Handle the tool call
+                # --- 5) Handle the tool call
                 tool_call = tool_calls[0]
                 function_name = tool_call.function.name
 
@@ -242,7 +327,6 @@ class LLM:
                     try:
                         # Call the actual tool function
                         result = fn(**function_args)
-
                         return result
 
                     except Exception as e:
@@ -264,6 +348,68 @@ class LLM:
                     logging.error(f"LiteLLM call failed: {str(e)}")
                 raise
 
+    def _format_messages_for_provider(self, messages: List[Dict[str, str]]) -> List[Dict[str, str]]:
+        """Format messages according to provider requirements.
+        
+        Args:
+            messages: List of message dictionaries with 'role' and 'content' keys.
+                     Can be empty or None.
+        
+        Returns:
+            List of formatted messages according to provider requirements.
+            For Anthropic models, ensures first message has 'user' role.
+        
+        Raises:
+            TypeError: If messages is None or contains invalid message format.
+        """
+        if messages is None:
+            raise TypeError("Messages cannot be None")
+            
+        # Validate message format first
+        for msg in messages:
+            if not isinstance(msg, dict) or "role" not in msg or "content" not in msg:
+                raise TypeError("Invalid message format. Each message must be a dict with 'role' and 'content' keys")
+            
+        if not self.is_anthropic:
+            return messages
+            
+        # Anthropic requires messages to start with 'user' role
+        if not messages or messages[0]["role"] == "system":
+            # If first message is system or empty, add a placeholder user message
+            return [{"role": "user", "content": "."}, *messages]
+                
+        return messages
+
+    def _get_custom_llm_provider(self) -> str:
+        """
+        Derives the custom_llm_provider from the model string.
+        - For example, if the model is "openrouter/deepseek/deepseek-chat", returns "openrouter".
+        - If the model is "gemini/gemini-1.5-pro", returns "gemini".
+        - If there is no '/', defaults to "openai".
+        """
+        if "/" in self.model:
+            return self.model.split("/")[0]
+        return "openai"
+
+    def _validate_call_params(self) -> None:
+        """
+        Validate parameters before making a call. Currently this only checks if
+        a response_format is provided and whether the model supports it.
+        The custom_llm_provider is dynamically determined from the model:
+          - E.g., "openrouter/deepseek/deepseek-chat" yields "openrouter"
+          - "gemini/gemini-1.5-pro" yields "gemini"
+          - If no slash is present, "openai" is assumed.
+        """
+        provider = self._get_custom_llm_provider()
+        if self.response_format is not None and not supports_response_schema(
+            model=self.model,
+            custom_llm_provider=provider,
+        ):
+            raise ValueError(
+                f"The model {self.model} does not support response_format for provider '{provider}'. "
+                "Please remove response_format or use a supported model."
+            )
+
     def supports_function_calling(self) -> bool:
         try:
             params = get_supported_openai_params(model=self.model)

src/crewai/memory/entity/entity_memory.py

@@ -1,3 +1,7 @@
+from typing import Optional
+
+from pydantic import PrivateAttr
+
 from crewai.memory.entity.entity_memory_item import EntityMemoryItem
 from crewai.memory.memory import Memory
 from crewai.memory.storage.rag_storage import RAGStorage
@@ -10,13 +14,15 @@ class EntityMemory(Memory):
     Inherits from the Memory class.
     """
 
-    def __init__(self, crew=None, embedder_config=None, storage=None, path=None):
-        if hasattr(crew, "memory_config") and crew.memory_config is not None:
-            self.memory_provider = crew.memory_config.get("provider")
-        else:
-            self.memory_provider = None
+    _memory_provider: Optional[str] = PrivateAttr()
 
-        if self.memory_provider == "mem0":
+    def __init__(self, crew=None, embedder_config=None, storage=None, path=None):
+        if crew and hasattr(crew, "memory_config") and crew.memory_config is not None:
+            memory_provider = crew.memory_config.get("provider")
+        else:
+            memory_provider = None
+
+        if memory_provider == "mem0":
             try:
                 from crewai.memory.storage.mem0_storage import Mem0Storage
             except ImportError:
@@ -36,11 +42,13 @@ class EntityMemory(Memory):
                     path=path,
                 )
             )
-        super().__init__(storage)
+
+        super().__init__(storage=storage)
+        self._memory_provider = memory_provider
 
     def save(self, item: EntityMemoryItem) -> None:  # type: ignore # BUG?: Signature of "save" incompatible with supertype "Memory"
         """Saves an entity item into the SQLite storage."""
-        if self.memory_provider == "mem0":
+        if self._memory_provider == "mem0":
             data = f"""
             Remember details about the following entity:
             Name: {item.name}

src/crewai/memory/long_term/long_term_memory.py

@@ -17,7 +17,7 @@ class LongTermMemory(Memory):
     def __init__(self, storage=None, path=None):
         if not storage:
             storage = LTMSQLiteStorage(db_path=path) if path else LTMSQLiteStorage()
-        super().__init__(storage)
+        super().__init__(storage=storage)
 
     def save(self, item: LongTermMemoryItem) -> None:  # type: ignore # BUG?: Signature of "save" incompatible with supertype "Memory"
         metadata = item.metadata

src/crewai/memory/memory.py

@@ -1,15 +1,19 @@
 from typing import Any, Dict, List, Optional
 
-from crewai.memory.storage.rag_storage import RAGStorage
+from pydantic import BaseModel
 
 
-class Memory:
+class Memory(BaseModel):
     """
     Base class for memory, now supporting agent tags and generic metadata.
     """
 
-    def __init__(self, storage: RAGStorage):
-        self.storage = storage
+    embedder_config: Optional[Dict[str, Any]] = None
+
+    storage: Any
+
+    def __init__(self, storage: Any, **data: Any):
+        super().__init__(storage=storage, **data)
 
     def save(
         self,

src/crewai/memory/short_term/short_term_memory.py

@@ -1,5 +1,7 @@
 from typing import Any, Dict, Optional
 
+from pydantic import PrivateAttr
+
 from crewai.memory.memory import Memory
 from crewai.memory.short_term.short_term_memory_item import ShortTermMemoryItem
 from crewai.memory.storage.rag_storage import RAGStorage
@@ -14,13 +16,15 @@ class ShortTermMemory(Memory):
     MemoryItem instances.
     """
 
-    def __init__(self, crew=None, embedder_config=None, storage=None, path=None):
-        if hasattr(crew, "memory_config") and crew.memory_config is not None:
-            self.memory_provider = crew.memory_config.get("provider")
-        else:
-            self.memory_provider = None
+    _memory_provider: Optional[str] = PrivateAttr()
 
-        if self.memory_provider == "mem0":
+    def __init__(self, crew=None, embedder_config=None, storage=None, path=None):
+        if crew and hasattr(crew, "memory_config") and crew.memory_config is not None:
+            memory_provider = crew.memory_config.get("provider")
+        else:
+            memory_provider = None
+
+        if memory_provider == "mem0":
             try:
                 from crewai.memory.storage.mem0_storage import Mem0Storage
             except ImportError:
@@ -39,7 +43,8 @@ class ShortTermMemory(Memory):
                     path=path,
                 )
             )
-        super().__init__(storage)
+        super().__init__(storage=storage)
+        self._memory_provider = memory_provider
 
     def save(
         self,
@@ -48,7 +53,7 @@ class ShortTermMemory(Memory):
         agent: Optional[str] = None,
     ) -> None:
         item = ShortTermMemoryItem(data=value, metadata=metadata, agent=agent)
-        if self.memory_provider == "mem0":
+        if self._memory_provider == "mem0":
             item.data = f"Remember the following insights from Agent run: {item.data}"
 
         super().save(value=item.data, metadata=item.metadata, agent=item.agent)

src/crewai/memory/storage/base_rag_storage.py

@@ -13,7 +13,7 @@ class BaseRAGStorage(ABC):
         self,
         type: str,
         allow_reset: bool = True,
-        embedder_config: Optional[Any] = None,
+        embedder_config: Optional[Dict[str, Any]] = None,
         crew: Any = None,
     ):
         self.type = type

src/crewai/memory/storage/kickoff_task_outputs_storage.py

@@ -1,12 +1,17 @@
 import json
+import logging
 import sqlite3
+from pathlib import Path
 from typing import Any, Dict, List, Optional
 
 from crewai.task import Task
 from crewai.utilities import Printer
 from crewai.utilities.crew_json_encoder import CrewJSONEncoder
+from crewai.utilities.errors import DatabaseError, DatabaseOperationError
 from crewai.utilities.paths import db_storage_path
 
+logger = logging.getLogger(__name__)
+
 
 class KickoffTaskOutputsSQLiteStorage:
     """
@@ -14,15 +19,24 @@ class KickoffTaskOutputsSQLiteStorage:
     """
 
     def __init__(
-        self, db_path: str = f"{db_storage_path()}/latest_kickoff_task_outputs.db"
+        self, db_path: Optional[str] = None
     ) -> None:
+        if db_path is None:
+            # Get the parent directory of the default db path and create our db file there
+            db_path = str(Path(db_storage_path()) / "latest_kickoff_task_outputs.db")
         self.db_path = db_path
         self._printer: Printer = Printer()
         self._initialize_db()
 
-    def _initialize_db(self):
-        """
-        Initializes the SQLite database and creates LTM table
+    def _initialize_db(self) -> None:
+        """Initialize the SQLite database and create the latest_kickoff_task_outputs table.
+
+        This method sets up the database schema for storing task outputs. It creates
+        a table with columns for task_id, expected_output, output (as JSON),
+        task_index, inputs (as JSON), was_replayed flag, and timestamp.
+
+        Raises:
+            DatabaseOperationError: If database initialization fails due to SQLite errors.
         """
         try:
             with sqlite3.connect(self.db_path) as conn:
@@ -43,10 +57,9 @@ class KickoffTaskOutputsSQLiteStorage:
 
                 conn.commit()
         except sqlite3.Error as e:
-            self._printer.print(
-                content=f"SAVING KICKOFF TASK OUTPUTS ERROR: An error occurred during database initialization: {e}",
-                color="red",
-            )
+            error_msg = DatabaseError.format_error(DatabaseError.INIT_ERROR, e)
+            logger.error(error_msg)
+            raise DatabaseOperationError(error_msg, e)
 
     def add(
         self,
@@ -55,9 +68,22 @@ class KickoffTaskOutputsSQLiteStorage:
         task_index: int,
         was_replayed: bool = False,
         inputs: Dict[str, Any] = {},
-    ):
+    ) -> None:
+        """Add a new task output record to the database.
+
+        Args:
+            task: The Task object containing task details.
+            output: Dictionary containing the task's output data.
+            task_index: Integer index of the task in the sequence.
+            was_replayed: Boolean indicating if this was a replay execution.
+            inputs: Dictionary of input parameters used for the task.
+
+        Raises:
+            DatabaseOperationError: If saving the task output fails due to SQLite errors.
+        """
         try:
             with sqlite3.connect(self.db_path) as conn:
+                conn.execute("BEGIN TRANSACTION")
                 cursor = conn.cursor()
                 cursor.execute(
                     """
@@ -76,21 +102,31 @@ class KickoffTaskOutputsSQLiteStorage:
                 )
                 conn.commit()
         except sqlite3.Error as e:
-            self._printer.print(
-                content=f"SAVING KICKOFF TASK OUTPUTS ERROR: An error occurred during database initialization: {e}",
-                color="red",
-            )
+            error_msg = DatabaseError.format_error(DatabaseError.SAVE_ERROR, e)
+            logger.error(error_msg)
+            raise DatabaseOperationError(error_msg, e)
 
     def update(
         self,
         task_index: int,
-        **kwargs,
-    ):
-        """
-        Updates an existing row in the latest_kickoff_task_outputs table based on task_index.
+        **kwargs: Any,
+    ) -> None:
+        """Update an existing task output record in the database.
+
+        Updates fields of a task output record identified by task_index. The fields
+        to update are provided as keyword arguments.
+
+        Args:
+            task_index: Integer index of the task to update.
+            **kwargs: Arbitrary keyword arguments representing fields to update.
+                     Values that are dictionaries will be JSON encoded.
+
+        Raises:
+            DatabaseOperationError: If updating the task output fails due to SQLite errors.
         """
         try:
             with sqlite3.connect(self.db_path) as conn:
+                conn.execute("BEGIN TRANSACTION")
                 cursor = conn.cursor()
 
                 fields = []
@@ -110,14 +146,23 @@ class KickoffTaskOutputsSQLiteStorage:
                 conn.commit()
 
                 if cursor.rowcount == 0:
-                    self._printer.print(
-                        f"No row found with task_index {task_index}. No update performed.",
-                        color="red",
-                    )
+                    logger.warning(f"No row found with task_index {task_index}. No update performed.")
         except sqlite3.Error as e:
-            self._printer.print(f"UPDATE KICKOFF TASK OUTPUTS ERROR: {e}", color="red")
+            error_msg = DatabaseError.format_error(DatabaseError.UPDATE_ERROR, e)
+            logger.error(error_msg)
+            raise DatabaseOperationError(error_msg, e)
 
-    def load(self) -> Optional[List[Dict[str, Any]]]:
+    def load(self) -> List[Dict[str, Any]]:
+        """Load all task output records from the database.
+
+        Returns:
+            List of dictionaries containing task output records, ordered by task_index.
+            Each dictionary contains: task_id, expected_output, output, task_index,
+            inputs, was_replayed, and timestamp.
+
+        Raises:
+            DatabaseOperationError: If loading task outputs fails due to SQLite errors.
+        """
         try:
             with sqlite3.connect(self.db_path) as conn:
                 cursor = conn.cursor()
@@ -144,23 +189,26 @@ class KickoffTaskOutputsSQLiteStorage:
                 return results
 
         except sqlite3.Error as e:
-            self._printer.print(
-                content=f"LOADING KICKOFF TASK OUTPUTS ERROR: An error occurred while querying kickoff task outputs: {e}",
-                color="red",
-            )
-            return None
+            error_msg = DatabaseError.format_error(DatabaseError.LOAD_ERROR, e)
+            logger.error(error_msg)
+            raise DatabaseOperationError(error_msg, e)
 
-    def delete_all(self):
-        """
-        Deletes all rows from the latest_kickoff_task_outputs table.
+    def delete_all(self) -> None:
+        """Delete all task output records from the database.
+
+        This method removes all records from the latest_kickoff_task_outputs table.
+        Use with caution as this operation cannot be undone.
+
+        Raises:
+            DatabaseOperationError: If deleting task outputs fails due to SQLite errors.
         """
         try:
             with sqlite3.connect(self.db_path) as conn:
+                conn.execute("BEGIN TRANSACTION")
                 cursor = conn.cursor()
                 cursor.execute("DELETE FROM latest_kickoff_task_outputs")
                 conn.commit()
         except sqlite3.Error as e:
-            self._printer.print(
-                content=f"ERROR: Failed to delete all kickoff task outputs: {e}",
-                color="red",
-            )
+            error_msg = DatabaseError.format_error(DatabaseError.DELETE_ERROR, e)
+            logger.error(error_msg)
+            raise DatabaseOperationError(error_msg, e)

src/crewai/memory/storage/ltm_sqlite_storage.py

@@ -1,5 +1,6 @@
 import json
 import sqlite3
+from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
 from crewai.utilities import Printer
@@ -12,10 +13,15 @@ class LTMSQLiteStorage:
     """
 
     def __init__(
-        self, db_path: str = f"{db_storage_path()}/long_term_memory_storage.db"
+        self, db_path: Optional[str] = None
     ) -> None:
+        if db_path is None:
+            # Get the parent directory of the default db path and create our db file there
+            db_path = str(Path(db_storage_path()) / "long_term_memory_storage.db")
         self.db_path = db_path
         self._printer: Printer = Printer()
+        # Ensure parent directory exists
+        Path(self.db_path).parent.mkdir(parents=True, exist_ok=True)
         self._initialize_db()
 
     def _initialize_db(self):

src/crewai/memory/storage/rag_storage.py

@@ -6,12 +6,17 @@ import shutil
 import uuid
 from typing import Any, Dict, List, Optional
 
-from chromadb.api import ClientAPI
+from chromadb.api import ClientAPI, Collection
+from chromadb.api.types import Documents, Embeddings, Metadatas
 
 from crewai.memory.storage.base_rag_storage import BaseRAGStorage
 from crewai.utilities import EmbeddingConfigurator
 from crewai.utilities.constants import MAX_FILE_NAME_LENGTH
 from crewai.utilities.paths import db_storage_path
+from crewai.utilities.exceptions.embedding_exceptions import (
+    EmbeddingConfigurationError,
+    EmbeddingInitializationError
+)
 
 
 @contextlib.contextmanager
@@ -32,15 +37,24 @@ def suppress_logging(
 
 
 class RAGStorage(BaseRAGStorage):
-    """
-    Extends Storage to handle embeddings for memory entries, improving
-    search efficiency.
+    """RAG-based Storage implementation using ChromaDB for vector storage and retrieval.
+
+    This class extends BaseRAGStorage to handle embeddings for memory entries,
+    improving search efficiency through vector similarity.
+
+    Attributes:
+        app: ChromaDB client instance
+        collection: ChromaDB collection for storing embeddings
+        type: Type of memory storage
+        allow_reset: Whether memory reset is allowed
+        path: Custom storage path for the database
     """
 
     app: ClientAPI | None = None
+    collection: Any = None
 
     def __init__(
-        self, type, allow_reset=True, embedder_config=None, crew=None, path=None
+        self, type: str, allow_reset: bool = True, embedder_config: Dict[str, Any] | None = None, crew: Any = None, path: str | None = None
     ):
         super().__init__(type, allow_reset, embedder_config, crew)
         agents = crew.agents if crew else []
@@ -50,7 +64,6 @@ class RAGStorage(BaseRAGStorage):
         self.storage_file_name = self._build_storage_file_name(type, agents)
 
         self.type = type
-
         self.allow_reset = allow_reset
         self.path = path
         self._initialize_app()
@@ -59,26 +72,36 @@ class RAGStorage(BaseRAGStorage):
         configurator = EmbeddingConfigurator()
         self.embedder_config = configurator.configure_embedder(self.embedder_config)
 
-    def _initialize_app(self):
+    def _initialize_app(self) -> None:
+        """Initialize the ChromaDB client and collection.
+
+        Raises:
+            RuntimeError: If ChromaDB client initialization fails
+            EmbeddingConfigurationError: If embedding configuration is invalid
+            EmbeddingInitializationError: If embedding function fails to initialize
+        """
         import chromadb
         from chromadb.config import Settings
 
         self._set_embedder_config()
-        chroma_client = chromadb.PersistentClient(
-            path=self.path if self.path else self.storage_file_name,
-            settings=Settings(allow_reset=self.allow_reset),
-        )
-
-        self.app = chroma_client
-
         try:
-            self.collection = self.app.get_collection(
-                name=self.type, embedding_function=self.embedder_config
-            )
-        except Exception:
-            self.collection = self.app.create_collection(
-                name=self.type, embedding_function=self.embedder_config
+            self.app = chromadb.PersistentClient(
+                path=self.path if self.path else self.storage_file_name,
+                settings=Settings(allow_reset=self.allow_reset),
             )
+            if not self.app:
+                raise RuntimeError("Failed to initialize ChromaDB client")
+
+            try:
+                self.collection = self.app.get_collection(
+                    name=self.type, embedding_function=self.embedder_config
+                )
+            except Exception:
+                self.collection = self.app.create_collection(
+                    name=self.type, embedding_function=self.embedder_config
+                )
+        except Exception as e:
+            raise RuntimeError(f"Failed to initialize ChromaDB: {str(e)}")
 
     def _sanitize_role(self, role: str) -> str:
         """
@@ -101,12 +124,21 @@ class RAGStorage(BaseRAGStorage):
         return f"{base_path}/{file_name}"
 
     def save(self, value: Any, metadata: Dict[str, Any]) -> None:
+        """Save a value with metadata to the memory storage.
+
+        Args:
+            value: The text content to store
+            metadata: Additional metadata for the stored content
+
+        Raises:
+            EmbeddingInitializationError: If embedding generation fails
+        """
         if not hasattr(self, "app") or not hasattr(self, "collection"):
             self._initialize_app()
         try:
             self._generate_embedding(value, metadata)
         except Exception as e:
-            logging.error(f"Error during {self.type} save: {str(e)}")
+            raise EmbeddingInitializationError(self.type, str(e))
 
     def search(
         self,
@@ -114,7 +146,18 @@ class RAGStorage(BaseRAGStorage):
         limit: int = 3,
         filter: Optional[dict] = None,
         score_threshold: float = 0.35,
-    ) -> List[Any]:
+    ) -> List[Dict[str, Any]]:
+        """Search for similar content in memory.
+
+        Args:
+            query: The search query text
+            limit: Maximum number of results to return
+            filter: Optional filter criteria
+            score_threshold: Minimum similarity score threshold
+
+        Returns:
+            List of matching results with metadata and scores
+        """
         if not hasattr(self, "app"):
             self._initialize_app()
 
@@ -138,37 +181,50 @@ class RAGStorage(BaseRAGStorage):
             logging.error(f"Error during {self.type} search: {str(e)}")
             return []
 
-    def _generate_embedding(self, text: str, metadata: Dict[str, Any]) -> None:  # type: ignore
+    def _generate_embedding(self, text: str, metadata: Optional[Dict[str, Any]] = None) -> Any:
+        """Generate and store embeddings for the given text.
+
+        Args:
+            text: The text to generate embeddings for
+            metadata: Optional additional metadata to store with the embeddings
+
+        Returns:
+            Any: The generated embedding or None if only storing
+        """
         if not hasattr(self, "app") or not hasattr(self, "collection"):
             self._initialize_app()
 
-        self.collection.add(
-            documents=[text],
-            metadatas=[metadata or {}],
-            ids=[str(uuid.uuid4())],
-        )
+        try:
+            self.collection.add(
+                documents=[text],
+                metadatas=[metadata or {}],
+                ids=[str(uuid.uuid4())],
+            )
+            return None
+        except Exception as e:
+            raise EmbeddingInitializationError(self.type, f"Failed to generate embedding: {str(e)}")
 
     def reset(self) -> None:
+        """Reset the memory storage by clearing the database and removing files.
+
+        Raises:
+            RuntimeError: If memory reset fails and allow_reset is False
+            EmbeddingConfigurationError: If embedding configuration is invalid during reinitialization
+        """
         try:
             if self.app:
                 self.app.reset()
-                shutil.rmtree(f"{db_storage_path()}/{self.type}")
+                storage_path = self.path if self.path else db_storage_path()
+                db_dir = os.path.join(storage_path, self.type)
+                if os.path.exists(db_dir):
+                    shutil.rmtree(db_dir)
                 self.app = None
                 self.collection = None
         except Exception as e:
             if "attempt to write a readonly database" in str(e):
-                # Ignore this specific error
+                # Ignore this specific error as it's expected in some environments
                 pass
             else:
-                raise Exception(
-                    f"An error occurred while resetting the {self.type} memory: {e}"
-                )
-
-    def _create_default_embedding_function(self):
-        from chromadb.utils.embedding_functions.openai_embedding_function import (
-            OpenAIEmbeddingFunction,
-        )
-
-        return OpenAIEmbeddingFunction(
-            api_key=os.getenv("OPENAI_API_KEY"), model_name="text-embedding-3-small"
-        )
+                if not self.allow_reset:
+                    raise RuntimeError(f"Failed to reset {self.type} memory: {str(e)}")
+                logging.error(f"Error during {self.type} memory reset: {str(e)}")

src/crewai/task.py

@@ -9,11 +9,13 @@ from copy import copy
 from hashlib import md5
 from pathlib import Path
 from typing import (
+    AbstractSet,
     Any,
     Callable,
     ClassVar,
     Dict,
     List,
+    Mapping,
     Optional,
     Set,
     Tuple,
@@ -109,7 +111,7 @@ class Task(BaseModel):
         description="Task output, it's final result after being executed", default=None
     )
     tools: Optional[List[BaseTool]] = Field(
-        default_factory=list,
+        default_factory=lambda: [],
         description="Tools the agent is limited to use for this task.",
     )
     id: UUID4 = Field(
@@ -125,7 +127,7 @@ class Task(BaseModel):
         description="A converter class used to export structured output",
         default=None,
     )
-    processed_by_agents: Set[str] = Field(default_factory=set)
+    processed_by_agents: Set[str] = Field(default_factory=lambda: set())
     guardrail: Optional[Callable[[TaskOutput], Tuple[bool, Any]]] = Field(
         default=None,
         description="Function to validate task output before proceeding to next task",
@@ -423,6 +425,10 @@ class Task(BaseModel):
         if self.callback:
             self.callback(self.output)
 
+        crew = self.agent.crew  # type: ignore[union-attr]
+        if crew and crew.task_callback and crew.task_callback != self.callback:
+            crew.task_callback(self.output)
+
         if self._execution_span:
             self._telemetry.task_ended(self._execution_span, self, agent.crew)
             self._execution_span = None
@@ -431,7 +437,9 @@ class Task(BaseModel):
             content = (
                 json_output
                 if json_output
-                else pydantic_output.model_dump_json() if pydantic_output else result
+                else pydantic_output.model_dump_json()
+                if pydantic_output
+                else result
             )
             self._save_file(content)
 
@@ -452,7 +460,7 @@ class Task(BaseModel):
         return "\n".join(tasks_slices)
 
     def interpolate_inputs_and_add_conversation_history(
-        self, inputs: Dict[str, Union[str, int, float]]
+        self, inputs: Dict[str, Union[str, int, float, Dict[str, Any], List[Any]]]
     ) -> None:
         """Interpolate inputs into the task description, expected output, and output file path.
            Add conversation history if present.
@@ -524,7 +532,9 @@ class Task(BaseModel):
             )
 
     def interpolate_only(
-        self, input_string: Optional[str], inputs: Dict[str, Union[str, int, float]]
+        self,
+        input_string: Optional[str],
+        inputs: Dict[str, Union[str, int, float, Dict[str, Any], List[Any]]],
     ) -> str:
         """Interpolate placeholders (e.g., {key}) in a string while leaving JSON untouched.
 
@@ -532,17 +542,39 @@ class Task(BaseModel):
             input_string: The string containing template variables to interpolate.
                          Can be None or empty, in which case an empty string is returned.
             inputs: Dictionary mapping template variables to their values.
-                   Supported value types are strings, integers, and floats.
-                   If input_string is empty or has no placeholders, inputs can be empty.
+                   Supported value types are strings, integers, floats, and dicts/lists
+                   containing only these types and other nested dicts/lists.
 
         Returns:
             The interpolated string with all template variables replaced with their values.
             Empty string if input_string is None or empty.
 
         Raises:
-            ValueError: If a required template variable is missing from inputs.
-            KeyError: If a template variable is not found in the inputs dictionary.
+            ValueError: If a value contains unsupported types
         """
+
+        # Validation function for recursive type checking
+        def validate_type(value: Any) -> None:
+            if value is None:
+                return
+            if isinstance(value, (str, int, float, bool)):
+                return
+            if isinstance(value, (dict, list)):
+                for item in value.values() if isinstance(value, dict) else value:
+                    validate_type(item)
+                return
+            raise ValueError(
+                f"Unsupported type {type(value).__name__} in inputs. "
+                "Only str, int, float, bool, dict, and list are allowed."
+            )
+
+        # Validate all input values
+        for key, value in inputs.items():
+            try:
+                validate_type(value)
+            except ValueError as e:
+                raise ValueError(f"Invalid value for key '{key}': {str(e)}") from e
+
         if input_string is None or not input_string:
             return ""
         if "{" not in input_string and "}" not in input_string:
@@ -551,15 +583,7 @@ class Task(BaseModel):
             raise ValueError(
                 "Inputs dictionary cannot be empty when interpolating variables"
             )
-
         try:
-            # Validate input types
-            for key, value in inputs.items():
-                if not isinstance(value, (str, int, float)):
-                    raise ValueError(
-                        f"Value for key '{key}' must be a string, integer, or float, got {type(value).__name__}"
-                    )
-
             escaped_string = input_string.replace("{", "{{").replace("}", "}}")
 
             for key in inputs.keys():
@@ -584,37 +608,56 @@ class Task(BaseModel):
         self.delegations += 1
 
     def copy(
+        self,
+        *,
+        include: Optional[AbstractSet[int] | AbstractSet[str] | Mapping[int, Any] | Mapping[str, Any]] = None,
+        exclude: Optional[AbstractSet[int] | AbstractSet[str] | Mapping[int, Any] | Mapping[str, Any]] = None,
+        update: Optional[Dict[str, Any]] = None,
+        deep: bool = False,
+    ) -> "Task":
+        """Create a copy of the Task."""
+        exclude_set = {"id", "agent", "context", "tools"}
+        if exclude:
+            if isinstance(exclude, (AbstractSet, set)):
+                exclude_set.update(str(x) for x in exclude)
+            elif isinstance(exclude, Mapping):
+                exclude_set.update(str(x) for x in exclude.keys())
+        
+        copied_task = super().copy(
+            include=include,
+            exclude=exclude_set,
+            update=update,
+            deep=deep,
+        )
+        
+        copied_task.id = uuid.uuid4()
+        copied_task.agent = None
+        copied_task.context = None
+        copied_task.tools = []
+        
+        return copied_task
+
+    def copy_with_agents(
         self, agents: List["BaseAgent"], task_mapping: Dict[str, "Task"]
     ) -> "Task":
-        """Create a deep copy of the Task."""
-        exclude = {
-            "id",
-            "agent",
-            "context",
-            "tools",
-        }
-
-        copied_data = self.model_dump(exclude=exclude)
-        copied_data = {k: v for k, v in copied_data.items() if v is not None}
-
-        cloned_context = (
-            [task_mapping[context_task.key] for context_task in self.context]
-            if self.context
-            else None
-        )
+        """Create a copy of the Task with agent references."""
+        copied_task = self.copy()
 
         def get_agent_by_role(role: str) -> Union["BaseAgent", None]:
             return next((agent for agent in agents if agent.role == role), None)
 
-        cloned_agent = get_agent_by_role(self.agent.role) if self.agent else None
-        cloned_tools = copy(self.tools) if self.tools else []
+        if self.agent:
+            copied_task.agent = get_agent_by_role(self.agent.role)
 
-        copied_task = Task(
-            **copied_data,
-            context=cloned_context,
-            agent=cloned_agent,
-            tools=cloned_tools,
-        )
+        if self.context:
+            copied_task.context = [
+                task_mapping[context_task.key]
+                for context_task in self.context
+                if context_task.key in task_mapping
+            ]
+
+        if self.tools:
+            copied_task.tools = copy(self.tools)
 
         return copied_task
 
@@ -652,19 +695,32 @@ class Task(BaseModel):
             return OutputFormat.PYDANTIC
         return OutputFormat.RAW
 
-    def _save_file(self, result: Any) -> None:
+    def _save_file(self, result: Union[Dict, str, Any]) -> None:
         """Save task output to a file.
 
+        Note:
+            For cross-platform file writing, especially on Windows, consider using FileWriterTool
+            from the crewai_tools package:
+                pip install 'crewai[tools]'
+                from crewai_tools import FileWriterTool
+
         Args:
             result: The result to save to the file. Can be a dict or any stringifiable object.
 
         Raises:
             ValueError: If output_file is not set
-            RuntimeError: If there is an error writing to the file
+            RuntimeError: If there is an error writing to the file. For cross-platform
+                compatibility, especially on Windows, use FileWriterTool from crewai_tools
+                package.
         """
         if self.output_file is None:
             raise ValueError("output_file is not set.")
 
+        FILEWRITER_RECOMMENDATION = (
+            "For cross-platform file writing, especially on Windows, "
+            "use FileWriterTool from crewai_tools package."
+        )
+
         try:
             resolved_path = Path(self.output_file).expanduser().resolve()
             directory = resolved_path.parent
@@ -680,7 +736,12 @@ class Task(BaseModel):
                 else:
                     file.write(str(result))
         except (OSError, IOError) as e:
-            raise RuntimeError(f"Failed to save output file: {e}")
+            raise RuntimeError(
+                "\n".join([
+                    f"Failed to save output file: {e}",
+                    FILEWRITER_RECOMMENDATION
+                ])
+            )
         return None
 
     def __repr__(self):

src/crewai/tasks/task_output.py

@@ -1,7 +1,9 @@
 import json
-from typing import Any, Dict, Optional
+from typing import Any, Callable, Dict, Optional
 
 from pydantic import BaseModel, Field, model_validator
+from pydantic.main import IncEx
+from typing_extensions import Literal
 
 from crewai.tasks.output_format import OutputFormat
 
@@ -34,8 +36,8 @@ class TaskOutput(BaseModel):
         self.summary = f"{excerpt}..."
         return self
 
-    @property
-    def json(self) -> Optional[str]:
+    def model_json(self) -> str:
+        """Get the JSON representation of the output."""
         if self.output_format != OutputFormat.JSON:
             raise ValueError(
                 """
@@ -44,8 +46,37 @@ class TaskOutput(BaseModel):
                 please make sure to set the output_json property for the task
                 """
             )
+        return json.dumps(self.json_dict) if self.json_dict else "{}"
 
-        return json.dumps(self.json_dict)
+    def model_dump_json(
+        self,
+        *,
+        indent: Optional[int] = None,
+        include: Optional[IncEx] = None,
+        exclude: Optional[IncEx] = None,
+        context: Optional[Any] = None,
+        by_alias: bool = False,
+        exclude_unset: bool = False,
+        exclude_defaults: bool = False,
+        exclude_none: bool = False,
+        round_trip: bool = False,
+        warnings: bool | Literal["none", "warn", "error"] = False,
+        serialize_as_any: bool = False,
+    ) -> str:
+        """Override model_dump_json to handle custom JSON output."""
+        return super().model_dump_json(
+            indent=indent,
+            include=include,
+            exclude=exclude,
+            context=context,
+            by_alias=by_alias,
+            exclude_unset=exclude_unset,
+            exclude_defaults=exclude_defaults,
+            exclude_none=exclude_none,
+            round_trip=round_trip,
+            warnings=warnings,
+            serialize_as_any=serialize_as_any,
+        )
 
     def to_dict(self) -> Dict[str, Any]:
         """Convert json_output and pydantic_output to a dictionary."""

src/crewai/tools/agent_tools/add_image_tool.py

@@ -7,11 +7,11 @@ from crewai.utilities import I18N
 
 i18n = I18N()
 
+
 class AddImageToolSchema(BaseModel):
     image_url: str = Field(..., description="The URL or path of the image to add")
     action: Optional[str] = Field(
-        default=None,
-        description="Optional context or question about the image"
+        default=None, description="Optional context or question about the image"
     )
 
 
@@ -36,10 +36,7 @@ class AddImageTool(BaseTool):
                 "image_url": {
                     "url": image_url,
                 },
-            }
+            },
         ]
 
-        return {
-            "role": "user",
-            "content": content
-        }
+        return {"role": "user", "content": content}

src/crewai/tools/agent_tools/base_agent_tools.py

@@ -82,12 +82,12 @@ class BaseAgentTool(BaseTool):
             available_agents = [agent.role for agent in self.agents]
             logger.debug(f"Available agents: {available_agents}")
 
-            agent = [  # type: ignore # Incompatible types in assignment (expression has type "list[BaseAgent]", variable has type "str | None")
+            matching_agents = [
                 available_agent
                 for available_agent in self.agents
                 if self.sanitize_agent_name(available_agent.role) == sanitized_name
             ]
-            logger.debug(f"Found {len(agent)} matching agents for role '{sanitized_name}'")
+            logger.debug(f"Found {len(matching_agents)} matching agents for role '{sanitized_name}'")
         except (AttributeError, ValueError) as e:
             # Handle specific exceptions that might occur during role name processing
             return self.i18n.errors("agent_tool_unexisting_coworker").format(
@@ -97,7 +97,7 @@ class BaseAgentTool(BaseTool):
                 error=str(e)
             )
 
-        if not agent:
+        if not matching_agents:
             # No matching agent found after sanitization
             return self.i18n.errors("agent_tool_unexisting_coworker").format(
                 coworkers="\n".join(
@@ -106,19 +106,19 @@ class BaseAgentTool(BaseTool):
                 error=f"No agent found with role '{sanitized_name}'"
             )
 
-        agent = agent[0]
+        selected_agent = matching_agents[0]
         try:
             task_with_assigned_agent = Task(
                 description=task,
-                agent=agent,
-                expected_output=agent.i18n.slice("manager_request"),
-                i18n=agent.i18n,
+                agent=selected_agent,
+                expected_output=selected_agent.i18n.slice("manager_request"),
+                i18n=selected_agent.i18n,
             )
-            logger.debug(f"Created task for agent '{self.sanitize_agent_name(agent.role)}': {task}")
-            return agent.execute_task(task_with_assigned_agent, context)
+            logger.debug(f"Created task for agent '{self.sanitize_agent_name(selected_agent.role)}': {task}")
+            return selected_agent.execute_task(task_with_assigned_agent, context)
         except Exception as e:
             # Handle task creation or execution errors
             return self.i18n.errors("agent_tool_execution_error").format(
-                agent_role=self.sanitize_agent_name(agent.role),
+                agent_role=self.sanitize_agent_name(selected_agent.role),
                 error=str(e)
             )

src/crewai/tools/base_tool.py

@@ -1,40 +1,36 @@
-import warnings
 from abc import ABC, abstractmethod
 from inspect import signature
-from typing import Any, Callable, Type, get_args, get_origin
+from typing import Any, Callable, Dict, Optional, Type, Tuple, get_args, get_origin
 
-from pydantic import (
-    BaseModel,
-    ConfigDict,
-    Field,
-    PydanticDeprecatedSince20,
-    create_model,
-    validator,
-)
+from pydantic import BaseModel, ConfigDict, Field, create_model, validator
+from pydantic.fields import FieldInfo
 from pydantic import BaseModel as PydanticBaseModel
 
 from crewai.tools.structured_tool import CrewStructuredTool
 
-# Ignore all "PydanticDeprecatedSince20" warnings globally
-warnings.filterwarnings("ignore", category=PydanticDeprecatedSince20)
-
+def _create_model_fields(fields: Dict[str, Tuple[Any, FieldInfo]]) -> Dict[str, Any]:
+    """Helper function to create model fields with proper type hints."""
+    return {name: (annotation, field) for name, (annotation, field) in fields.items()}
 
 class BaseTool(BaseModel, ABC):
+    """Base class for all tools."""
+
     class _ArgsSchemaPlaceholder(PydanticBaseModel):
         pass
 
-    model_config = ConfigDict()
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+    func: Optional[Callable] = None
 
     name: str
     """The unique name of the tool that clearly communicates its purpose."""
     description: str
     """Used to tell the model how/when/why to use the tool."""
-    args_schema: Type[PydanticBaseModel] = Field(default_factory=_ArgsSchemaPlaceholder)
+    args_schema: Type[PydanticBaseModel] = Field(default=_ArgsSchemaPlaceholder)
     """The schema for the arguments that the tool accepts."""
     description_updated: bool = False
     """Flag to check if the description has been updated."""
     cache_function: Callable = lambda _args=None, _result=None: True
-    """Function that will be used to determine if the tool should be cached, should return a boolean. If None, the tool will be cached."""
+    """Function that will be used to determine if the tool should be cached."""
     result_as_answer: bool = False
     """Flag to check if the tool should be the final agent answer."""
 
@@ -57,7 +53,6 @@ class BaseTool(BaseModel, ABC):
 
     def model_post_init(self, __context: Any) -> None:
         self._generate_description()
-
         super().model_post_init(__context)
 
     def run(
@@ -87,50 +82,7 @@ class BaseTool(BaseModel, ABC):
             result_as_answer=self.result_as_answer,
         )
 
-    @classmethod
-    def from_langchain(cls, tool: Any) -> "BaseTool":
-        """Create a Tool instance from a CrewStructuredTool.
-
-        This method takes a CrewStructuredTool object and converts it into a
-        Tool instance. It ensures that the provided tool has a callable 'func'
-        attribute and infers the argument schema if not explicitly provided.
-        """
-        if not hasattr(tool, "func") or not callable(tool.func):
-            raise ValueError("The provided tool must have a callable 'func' attribute.")
-
-        args_schema = getattr(tool, "args_schema", None)
-
-        if args_schema is None:
-            # Infer args_schema from the function signature if not provided
-            func_signature = signature(tool.func)
-            annotations = func_signature.parameters
-            args_fields = {}
-            for name, param in annotations.items():
-                if name != "self":
-                    param_annotation = (
-                        param.annotation if param.annotation != param.empty else Any
-                    )
-                    field_info = Field(
-                        default=...,
-                        description="",
-                    )
-                    args_fields[name] = (param_annotation, field_info)
-            if args_fields:
-                args_schema = create_model(f"{tool.name}Input", **args_fields)
-            else:
-                # Create a default schema with no fields if no parameters are found
-                args_schema = create_model(
-                    f"{tool.name}Input", __base__=PydanticBaseModel
-                )
-
-        return cls(
-            name=getattr(tool, "name", "Unnamed Tool"),
-            description=getattr(tool, "description", ""),
-            func=tool.func,
-            args_schema=args_schema,
-        )
-
-    def _set_args_schema(self):
+    def _set_args_schema(self) -> None:
         if self.args_schema is None:
             class_name = f"{self.__class__.__name__}Schema"
             self.args_schema = type(
@@ -145,7 +97,7 @@ class BaseTool(BaseModel, ABC):
                 },
             )
 
-    def _generate_description(self):
+    def _generate_description(self) -> None:
         args_schema = {
             name: {
                 "description": field.description,
@@ -179,79 +131,25 @@ class BaseTool(BaseModel, ABC):
 
 
 class Tool(BaseTool):
-    """The function that will be executed when the tool is called."""
+    """Tool class that wraps a function."""
 
     func: Callable
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+
+    def __init__(self, **kwargs):
+        if "func" not in kwargs:
+            raise ValueError("Tool requires a 'func' argument")
+        super().__init__(**kwargs)
 
     def _run(self, *args: Any, **kwargs: Any) -> Any:
         return self.func(*args, **kwargs)
 
-    @classmethod
-    def from_langchain(cls, tool: Any) -> "Tool":
-        """Create a Tool instance from a CrewStructuredTool.
 
-        This method takes a CrewStructuredTool object and converts it into a
-        Tool instance. It ensures that the provided tool has a callable 'func'
-        attribute and infers the argument schema if not explicitly provided.
-
-        Args:
-            tool (Any): The CrewStructuredTool object to be converted.
-
-        Returns:
-            Tool: A new Tool instance created from the provided CrewStructuredTool.
-
-        Raises:
-            ValueError: If the provided tool does not have a callable 'func' attribute.
-        """
-        if not hasattr(tool, "func") or not callable(tool.func):
-            raise ValueError("The provided tool must have a callable 'func' attribute.")
-
-        args_schema = getattr(tool, "args_schema", None)
-
-        if args_schema is None:
-            # Infer args_schema from the function signature if not provided
-            func_signature = signature(tool.func)
-            annotations = func_signature.parameters
-            args_fields = {}
-            for name, param in annotations.items():
-                if name != "self":
-                    param_annotation = (
-                        param.annotation if param.annotation != param.empty else Any
-                    )
-                    field_info = Field(
-                        default=...,
-                        description="",
-                    )
-                    args_fields[name] = (param_annotation, field_info)
-            if args_fields:
-                args_schema = create_model(f"{tool.name}Input", **args_fields)
-            else:
-                # Create a default schema with no fields if no parameters are found
-                args_schema = create_model(
-                    f"{tool.name}Input", __base__=PydanticBaseModel
-                )
-
-        return cls(
-            name=getattr(tool, "name", "Unnamed Tool"),
-            description=getattr(tool, "description", ""),
-            func=tool.func,
-            args_schema=args_schema,
-        )
-
-
-def to_langchain(
-    tools: list[BaseTool | CrewStructuredTool],
-) -> list[CrewStructuredTool]:
-    return [t.to_structured_tool() if isinstance(t, BaseTool) else t for t in tools]
-
-
-def tool(*args):
-    """
-    Decorator to create a tool from a function.
-    """
+def tool(*args: Any) -> Any:
+    """Decorator to create a tool from a function."""
 
     def _make_with_name(tool_name: str) -> Callable:
-        def _make_tool(f: Callable) -> BaseTool:
+        def _make_tool(f: Callable) -> Tool:
             if f.__doc__ is None:
                 raise ValueError("Function must have a docstring")
             if f.__annotations__ is None:

src/crewai/tools/structured_tool.py

@@ -2,9 +2,14 @@ from __future__ import annotations
 
 import inspect
 import textwrap
-from typing import Any, Callable, Optional, Union, get_type_hints
+from typing import Any, Callable, Dict, Optional, Tuple, Union, get_type_hints
 
-from pydantic import BaseModel, Field, create_model
+from pydantic import BaseModel, ConfigDict, Field, create_model
+from pydantic.fields import FieldInfo
+
+def _create_model_fields(fields: Dict[str, Tuple[Any, FieldInfo]]) -> Dict[str, Any]:
+    """Helper function to create model fields with proper type hints."""
+    return {name: (annotation, field) for name, (annotation, field) in fields.items()}
 
 from crewai.utilities.logger import Logger
 
@@ -142,7 +147,8 @@ class CrewStructuredTool:
 
         # Create model
         schema_name = f"{name.title()}Schema"
-        return create_model(schema_name, **fields)
+        model_fields = _create_model_fields(fields)
+        return create_model(schema_name, __base__=BaseModel, **model_fields)
 
     def _validate_function_signature(self) -> None:
         """Validate that the function signature matches the args schema."""

src/crewai/tools/tool_usage.py

@@ -1,12 +1,13 @@
 import ast
 import datetime
 import json
-import re
 import time
 from difflib import SequenceMatcher
+from json import JSONDecodeError
 from textwrap import dedent
-from typing import Any, Dict, List, Union
+from typing import Any, Dict, List, Optional, Union
 
+import json5
 from json_repair import repair_json
 
 import crewai.utilities.events as events
@@ -407,28 +408,55 @@ class ToolUsage:
                 )
             return self._tool_calling(tool_string)
 
-    def _validate_tool_input(self, tool_input: str) -> Dict[str, Any]:
+    def _validate_tool_input(self, tool_input: Optional[str]) -> Dict[str, Any]:
+        if tool_input is None:
+            return {}
+
+        if not isinstance(tool_input, str) or not tool_input.strip():
+            raise Exception(
+                "Tool input must be a valid dictionary in JSON or Python literal format"
+            )
+
+        # Attempt 1: Parse as JSON
         try:
-            # Replace Python literals with JSON equivalents
-            replacements = {
-                r"'": '"',
-                r"None": "null",
-                r"True": "true",
-                r"False": "false",
-            }
-            for pattern, replacement in replacements.items():
-                tool_input = re.sub(pattern, replacement, tool_input)
-
             arguments = json.loads(tool_input)
-        except json.JSONDecodeError:
-            # Attempt to repair JSON string
-            repaired_input = repair_json(tool_input)
-            try:
-                arguments = json.loads(repaired_input)
-            except json.JSONDecodeError as e:
-                raise Exception(f"Invalid tool input JSON: {e}")
+            if isinstance(arguments, dict):
+                return arguments
+        except (JSONDecodeError, TypeError):
+            pass  # Continue to the next parsing attempt
 
-        return arguments
+        # Attempt 2: Parse as Python literal
+        try:
+            arguments = ast.literal_eval(tool_input)
+            if isinstance(arguments, dict):
+                return arguments
+        except (ValueError, SyntaxError):
+            pass  # Continue to the next parsing attempt
+
+        # Attempt 3: Parse as JSON5
+        try:
+            arguments = json5.loads(tool_input)
+            if isinstance(arguments, dict):
+                return arguments
+        except (JSONDecodeError, ValueError, TypeError):
+            pass  # Continue to the next parsing attempt
+
+        # Attempt 4: Repair JSON
+        try:
+            repaired_input = repair_json(tool_input)
+            self._printer.print(
+                content=f"Repaired JSON: {repaired_input}", color="blue"
+            )
+            arguments = json.loads(repaired_input)
+            if isinstance(arguments, dict):
+                return arguments
+        except Exception as e:
+            self._printer.print(content=f"Failed to repair JSON: {e}", color="red")
+
+        # If all parsing attempts fail, raise an error
+        raise Exception(
+            "Tool input must be a valid dictionary in JSON or Python literal format"
+        )
 
     def on_tool_error(self, tool: Any, tool_calling: ToolCalling, e: Exception) -> None:
         event_data = self._prepare_event_data(tool, tool_calling)

src/crewai/translations/en.json

@@ -15,7 +15,7 @@
     "final_answer_format": "If you don't need to use any more tools, you must give your best complete final answer, make sure it satisfies the expected criteria, use the EXACT format below:\n\n```\nThought: I now can give a great answer\nFinal Answer: my best complete final answer to the task.\n\n```",
     "format_without_tools": "\nSorry, I didn't use the right format. I MUST either use a tool (among the available ones), OR give my best final answer.\nHere is the expected format I must follow:\n\n```\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n```\n This Thought/Action/Action Input/Result process can repeat N times. Once I know the final answer, I must return the following format:\n\n```\nThought: I now can give a great answer\nFinal Answer: Your final answer must be the great and the most complete as possible, it must be outcome described\n\n```",
     "task_with_context": "{task}\n\nThis is the context you're working with:\n{context}",
-    "expected_output": "\nThis is the expect criteria for your final answer: {expected_output}\nyou MUST return the actual complete content as the final answer, not a summary.",
+    "expected_output": "\nThis is the expected criteria for your final answer: {expected_output}\nyou MUST return the actual complete content as the final answer, not a summary.",
     "human_feedback": "You got human feedback on your work, re-evaluate it and give a new Final Answer when ready.\n {human_feedback}",
     "getting_input": "This is the agent's final answer: {final_answer}\n\n",
     "summarizer_system_message": "You are a helpful assistant that summarizes text.",
@@ -24,7 +24,8 @@
     "manager_request": "Your best answer to your coworker asking you this, accounting for the context shared.",
     "formatted_task_instructions": "Ensure your final answer contains only the content in the following format: {output_format}\n\nEnsure the final output does not include any code block markers like ```json or ```python.",
     "human_feedback_classification": "Determine if the following feedback indicates that the user is satisfied or if further changes are needed. Respond with 'True' if further changes are needed, or 'False' if the user is satisfied. **Important** Do not include any additional commentary outside of your 'True' or 'False' response.\n\nFeedback: \"{feedback}\"",
-    "conversation_history_instruction": "You are a member of a crew collaborating to achieve a common goal. Your task is a specific action that contributes to this larger objective. For additional context, please review the conversation history between you and the user that led to the initiation of this crew. Use any relevant information or feedback from the conversation to inform your task execution and ensure your response aligns with both the immediate task and the crew's overall goals."
+    "conversation_history_instruction": "You are a member of a crew collaborating to achieve a common goal. Your task is a specific action that contributes to this larger objective. For additional context, please review the conversation history between you and the user that led to the initiation of this crew. Use any relevant information or feedback from the conversation to inform your task execution and ensure your response aligns with both the immediate task and the crew's overall goals.",
+    "feedback_instructions": "User feedback: {feedback}\nInstructions: Use this feedback to enhance the next output iteration.\nNote: Do not respond or add commentary."
   },
   "errors": {
     "force_final_answer_error": "You can't keep going, here is the best final answer you generated:\n\n {formatted_answer}",

src/crewai/utilities/constants.py

@@ -4,3 +4,7 @@ DEFAULT_SCORE_THRESHOLD = 0.35
 KNOWLEDGE_DIRECTORY = "knowledge"
 MAX_LLM_RETRY = 3
 MAX_FILE_NAME_LENGTH = 255
+
+# Default embedding configuration
+DEFAULT_EMBEDDING_PROVIDER = "openai"
+DEFAULT_EMBEDDING_MODEL = "text-embedding-3-small"

src/crewai/utilities/converter.py

@@ -26,17 +26,24 @@ class Converter(OutputConverter):
             if self.llm.supports_function_calling():
                 return self._create_instructor().to_pydantic()
             else:
-                return self.llm.call(
+                response = self.llm.call(
                     [
                         {"role": "system", "content": self.instructions},
                         {"role": "user", "content": self.text},
                     ]
                 )
+                return self.model.model_validate_json(response)
+        except ValidationError as e:
+            if current_attempt < self.max_attempts:
+                return self.to_pydantic(current_attempt + 1)
+            raise ConverterError(
+                f"Failed to convert text into a Pydantic model due to the following validation error: {e}"
+            )
         except Exception as e:
             if current_attempt < self.max_attempts:
                 return self.to_pydantic(current_attempt + 1)
-            return ConverterError(
-                f"Failed to convert text into a pydantic model due to the following error: {e}"
+            raise ConverterError(
+                f"Failed to convert text into a Pydantic model due to the following error: {e}"
             )
 
     def to_json(self, current_attempt=1):
@@ -66,7 +73,6 @@ class Converter(OutputConverter):
             llm=self.llm,
             model=self.model,
             content=self.text,
-            instructions=self.instructions,
         )
         return inst
 
@@ -187,10 +193,15 @@ def convert_with_instructions(
 
 
 def get_conversion_instructions(model: Type[BaseModel], llm: Any) -> str:
-    instructions = "I'm gonna convert this raw text into valid JSON."
+    instructions = "Please convert the following text into valid JSON."
     if llm.supports_function_calling():
         model_schema = PydanticSchemaParser(model=model).get_schema()
-        instructions = f"{instructions}\n\nThe json should have the following structure, with the following keys:\n{model_schema}"
+        instructions += (
+            f"\n\nThe JSON should follow this schema:\n```json\n{model_schema}\n```"
+        )
+    else:
+        model_description = generate_model_description(model)
+        instructions += f"\n\nThe JSON should follow this format:\n{model_description}"
     return instructions
 
 
@@ -230,9 +241,13 @@ def generate_model_description(model: Type[BaseModel]) -> str:
         origin = get_origin(field_type)
         args = get_args(field_type)
 
-        if origin is Union and type(None) in args:
+        if origin is Union or (origin is None and len(args) > 0):
+            # Handle both Union and the new '|' syntax
             non_none_args = [arg for arg in args if arg is not type(None)]
-            return f"Optional[{describe_field(non_none_args[0])}]"
+            if len(non_none_args) == 1:
+                return f"Optional[{describe_field(non_none_args[0])}]"
+            else:
+                return f"Optional[Union[{', '.join(describe_field(arg) for arg in non_none_args)}]]"
         elif origin is list:
             return f"List[{describe_field(args[0])}]"
         elif origin is dict:
@@ -241,8 +256,10 @@ def generate_model_description(model: Type[BaseModel]) -> str:
             return f"Dict[{key_type}, {value_type}]"
         elif isinstance(field_type, type) and issubclass(field_type, BaseModel):
             return generate_model_description(field_type)
-        else:
+        elif hasattr(field_type, "__name__"):
             return field_type.__name__
+        else:
+            return str(field_type)
 
     fields = model.__annotations__
     field_descriptions = [

src/crewai/utilities/embedding_configurator.py

@@ -1,9 +1,15 @@
 import os
-from typing import Any, Dict, cast
+from typing import Any, Dict, List, Optional, cast
 
 from chromadb import Documents, EmbeddingFunction, Embeddings
 from chromadb.api.types import validate_embedding_function
 
+from crewai.utilities.exceptions.embedding_exceptions import (
+    EmbeddingConfigurationError,
+    EmbeddingProviderError,
+    EmbeddingInitializationError
+)
+
 
 class EmbeddingConfigurator:
     def __init__(self):
@@ -21,7 +27,7 @@ class EmbeddingConfigurator:
 
     def configure_embedder(
         self,
-        embedder_config: Dict[str, Any] | None = None,
+        embedder_config: Optional[Dict[str, Any]] = None,
     ) -> EmbeddingFunction:
         """Configures and returns an embedding function based on the provided config."""
         if embedder_config is None:
@@ -36,42 +42,47 @@ class EmbeddingConfigurator:
                 validate_embedding_function(provider)
                 return provider
             except Exception as e:
-                raise ValueError(f"Invalid custom embedding function: {str(e)}")
+                raise EmbeddingConfigurationError(f"Invalid custom embedding function: {str(e)}")
 
-        if provider not in self.embedding_functions:
-            raise Exception(
-                f"Unsupported embedding provider: {provider}, supported providers: {list(self.embedding_functions.keys())}"
-            )
+        if not provider or provider not in self.embedding_functions:
+            raise EmbeddingProviderError(str(provider), list(self.embedding_functions.keys()))
 
-        return self.embedding_functions[provider](config, model_name)
+        try:
+            return self.embedding_functions[str(provider)](config, model_name)
+        except Exception as e:
+            raise EmbeddingInitializationError(str(provider), str(e))
 
     @staticmethod
-    def _create_default_embedding_function():
-        from chromadb.utils.embedding_functions.openai_embedding_function import (
-            OpenAIEmbeddingFunction,
-        )
-
-        return OpenAIEmbeddingFunction(
-            api_key=os.getenv("OPENAI_API_KEY"), model_name="text-embedding-3-small"
-        )
+    def _create_default_embedding_function() -> EmbeddingFunction:
+        from crewai.utilities.constants import DEFAULT_EMBEDDING_PROVIDER, DEFAULT_EMBEDDING_MODEL
+        
+        provider = os.getenv("CREWAI_EMBEDDING_PROVIDER", DEFAULT_EMBEDDING_PROVIDER)
+        model = os.getenv("CREWAI_EMBEDDING_MODEL", DEFAULT_EMBEDDING_MODEL)
+        
+        if provider == "openai":
+            api_key = os.getenv("OPENAI_API_KEY")
+            if not api_key:
+                raise EmbeddingConfigurationError("OpenAI API key is required but not provided")
+            from chromadb.utils.embedding_functions.openai_embedding_function import OpenAIEmbeddingFunction
+            return OpenAIEmbeddingFunction(api_key=api_key, model_name=model)
+        elif provider == "ollama":
+            from chromadb.utils.embedding_functions.ollama_embedding_function import OllamaEmbeddingFunction
+            url = os.getenv("CREWAI_OLLAMA_URL", "http://localhost:11434/api/embeddings")
+            return OllamaEmbeddingFunction(url=url, model_name=model)
+        else:
+            raise EmbeddingProviderError(provider, ["openai", "ollama"])
 
     @staticmethod
-    def _configure_openai(config, model_name):
-        from chromadb.utils.embedding_functions.openai_embedding_function import (
-            OpenAIEmbeddingFunction,
-        )
-
+    def _configure_openai(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.openai_embedding_function import OpenAIEmbeddingFunction
         return OpenAIEmbeddingFunction(
             api_key=config.get("api_key") or os.getenv("OPENAI_API_KEY"),
             model_name=model_name,
         )
 
     @staticmethod
-    def _configure_azure(config, model_name):
-        from chromadb.utils.embedding_functions.openai_embedding_function import (
-            OpenAIEmbeddingFunction,
-        )
-
+    def _configure_azure(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.openai_embedding_function import OpenAIEmbeddingFunction
         return OpenAIEmbeddingFunction(
             api_key=config.get("api_key"),
             api_base=config.get("api_base"),
@@ -81,79 +92,62 @@ class EmbeddingConfigurator:
         )
 
     @staticmethod
-    def _configure_ollama(config, model_name):
-        from chromadb.utils.embedding_functions.ollama_embedding_function import (
-            OllamaEmbeddingFunction,
-        )
-
+    def _configure_ollama(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.ollama_embedding_function import OllamaEmbeddingFunction
         return OllamaEmbeddingFunction(
             url=config.get("url", "http://localhost:11434/api/embeddings"),
             model_name=model_name,
         )
 
     @staticmethod
-    def _configure_vertexai(config, model_name):
-        from chromadb.utils.embedding_functions.google_embedding_function import (
-            GoogleVertexEmbeddingFunction,
-        )
-
+    def _configure_vertexai(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.google_embedding_function import GoogleVertexEmbeddingFunction
         return GoogleVertexEmbeddingFunction(
             model_name=model_name,
             api_key=config.get("api_key"),
         )
 
     @staticmethod
-    def _configure_google(config, model_name):
-        from chromadb.utils.embedding_functions.google_embedding_function import (
-            GoogleGenerativeAiEmbeddingFunction,
-        )
-
+    def _configure_google(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.google_embedding_function import GoogleGenerativeAiEmbeddingFunction
         return GoogleGenerativeAiEmbeddingFunction(
             model_name=model_name,
             api_key=config.get("api_key"),
         )
 
     @staticmethod
-    def _configure_cohere(config, model_name):
-        from chromadb.utils.embedding_functions.cohere_embedding_function import (
-            CohereEmbeddingFunction,
-        )
-
+    def _configure_cohere(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.cohere_embedding_function import CohereEmbeddingFunction
         return CohereEmbeddingFunction(
             model_name=model_name,
             api_key=config.get("api_key"),
         )
 
     @staticmethod
-    def _configure_bedrock(config, model_name):
-        from chromadb.utils.embedding_functions.amazon_bedrock_embedding_function import (
-            AmazonBedrockEmbeddingFunction,
-        )
-
+    def _configure_bedrock(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.amazon_bedrock_embedding_function import AmazonBedrockEmbeddingFunction
         return AmazonBedrockEmbeddingFunction(
             session=config.get("session"),
         )
 
     @staticmethod
-    def _configure_huggingface(config, model_name):
-        from chromadb.utils.embedding_functions.huggingface_embedding_function import (
-            HuggingFaceEmbeddingServer,
-        )
-
+    def _configure_huggingface(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
+        from chromadb.utils.embedding_functions.huggingface_embedding_function import HuggingFaceEmbeddingServer
         return HuggingFaceEmbeddingServer(
             url=config.get("api_url"),
         )
 
     @staticmethod
-    def _configure_watson(config, model_name):
+    def _configure_watson(config: Dict[str, Any], model_name: str) -> EmbeddingFunction:
         try:
             import ibm_watsonx_ai.foundation_models as watson_models
             from ibm_watsonx_ai import Credentials
             from ibm_watsonx_ai.metanames import EmbedTextParamsMetaNames as EmbedParams
         except ImportError as e:
-            raise ImportError(
-                "IBM Watson dependencies are not installed. Please install them to use Watson embedding."
-            ) from e
+            raise EmbeddingConfigurationError(
+                "IBM Watson dependencies are not installed. Please install them to use Watson embedding.",
+                provider="watson"
+            )
 
         class WatsonEmbeddingFunction(EmbeddingFunction):
             def __call__(self, input: Documents) -> Embeddings:
@@ -178,7 +172,6 @@ class EmbeddingConfigurator:
                     embeddings = embedding.embed_documents(input)
                     return cast(Embeddings, embeddings)
                 except Exception as e:
-                    print("Error during Watson embedding:", e)
-                    raise e
+                    raise EmbeddingInitializationError("watson", str(e))
 
         return WatsonEmbeddingFunction()

src/crewai/utilities/errors.py

@@ -0,0 +1,39 @@
+"""Error message definitions for CrewAI database operations."""
+from typing import Optional
+
+
+class DatabaseOperationError(Exception):
+    """Base exception class for database operation errors."""
+
+    def __init__(self, message: str, original_error: Optional[Exception] = None):
+        """Initialize the database operation error.
+
+        Args:
+            message: The error message to display
+            original_error: The original exception that caused this error, if any
+        """
+        super().__init__(message)
+        self.original_error = original_error
+
+
+class DatabaseError:
+    """Standardized error message templates for database operations."""
+
+    INIT_ERROR: str = "Database initialization error: {}"
+    SAVE_ERROR: str = "Error saving task outputs: {}"
+    UPDATE_ERROR: str = "Error updating task outputs: {}"
+    LOAD_ERROR: str = "Error loading task outputs: {}"
+    DELETE_ERROR: str = "Error deleting task outputs: {}"
+
+    @classmethod
+    def format_error(cls, template: str, error: Exception) -> str:
+        """Format an error message with the given template and error.
+
+        Args:
+            template: The error message template to use
+            error: The exception to format into the template
+
+        Returns:
+            The formatted error message
+        """
+        return template.format(str(error))

src/crewai/utilities/evaluators/task_evaluator.py

@@ -92,13 +92,34 @@ class TaskEvaluator:
         """
 
         output_training_data = training_data[agent_id]
-
         final_aggregated_data = ""
-        for _, data in output_training_data.items():
+
+        for iteration, data in output_training_data.items():
+            improved_output = data.get("improved_output")
+            initial_output = data.get("initial_output")
+            human_feedback = data.get("human_feedback")
+
+            if not all([improved_output, initial_output, human_feedback]):
+                missing_fields = [
+                    field
+                    for field in ["improved_output", "initial_output", "human_feedback"]
+                    if not data.get(field)
+                ]
+                error_msg = (
+                    f"Critical training data error: Missing fields ({', '.join(missing_fields)}) "
+                    f"for agent {agent_id} in iteration {iteration}.\n"
+                    "This indicates a broken training process. "
+                    "Cannot proceed with evaluation.\n"
+                    "Please check your training implementation."
+                )
+                raise ValueError(error_msg)
+
             final_aggregated_data += (
-                f"Initial Output:\n{data['initial_output']}\n\n"
-                f"Human Feedback:\n{data['human_feedback']}\n\n"
-                f"Improved Output:\n{data['improved_output']}\n\n"
+                f"Iteration: {iteration}\n"
+                f"Initial Output:\n{initial_output}\n\n"
+                f"Human Feedback:\n{human_feedback}\n\n"
+                f"Improved Output:\n{improved_output}\n\n"
+                "------------------------------------------------\n\n"
             )
 
         evaluation_query = (

src/crewai/utilities/exceptions/embedding_exceptions.py

@@ -0,0 +1,20 @@
+from typing import List, Optional
+
+
+class EmbeddingConfigurationError(Exception):
+    def __init__(self, message: str, provider: Optional[str] = None):
+        self.message = message
+        self.provider = provider
+        super().__init__(self.message)
+
+
+class EmbeddingProviderError(EmbeddingConfigurationError):
+    def __init__(self, provider: str, supported_providers: List[str]):
+        message = f"Unsupported embedding provider: {provider}, supported providers: {supported_providers}"
+        super().__init__(message, provider)
+
+
+class EmbeddingInitializationError(EmbeddingConfigurationError):
+    def __init__(self, provider: str, error: str):
+        message = f"Failed to initialize embedding function for provider {provider}: {error}"
+        super().__init__(message, provider)

src/crewai/utilities/file_handler.py

@@ -1,30 +1,64 @@
+import json
 import os
 import pickle
 from datetime import datetime
+from typing import Union
 
 
 class FileHandler:
-    """take care of file operations, currently it only logs messages to a file"""
+    """Handler for file operations supporting both JSON and text-based logging.
+    
+    Args:
+        file_path (Union[bool, str]): Path to the log file or boolean flag
+    """
 
-    def __init__(self, file_path):
-        if isinstance(file_path, bool):
+    def __init__(self, file_path: Union[bool, str]):
+        self._initialize_path(file_path)
+        
+    def _initialize_path(self, file_path: Union[bool, str]):
+        if file_path is True:  # File path is boolean True
             self._path = os.path.join(os.curdir, "logs.txt")
-        elif isinstance(file_path, str):
-            self._path = file_path
+        
+        elif isinstance(file_path, str):  # File path is a string
+            if file_path.endswith((".json", ".txt")):
+                self._path = file_path  # No modification if the file ends with .json or .txt
+            else:
+                self._path = file_path + ".txt"  # Append .txt if the file doesn't end with .json or .txt
+        
         else:
-            raise ValueError("file_path must be either a boolean or a string.")
-
+            raise ValueError("file_path must be a string or boolean.")  # Handle the case where file_path isn't valid
+        
     def log(self, **kwargs):
-        now = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-        message = (
-            f"{now}: "
-            + ", ".join([f'{key}="{value}"' for key, value in kwargs.items()])
-            + "\n"
-        )
-        with open(self._path, "a", encoding="utf-8") as file:
-            file.write(message + "\n")
+        try:
+            now = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+            log_entry = {"timestamp": now, **kwargs}
 
+            if self._path.endswith(".json"):
+                # Append log in JSON format
+                with open(self._path, "a", encoding="utf-8") as file:
+                    # If the file is empty, start with a list; else, append to it
+                    try:
+                        # Try reading existing content to avoid overwriting
+                        with open(self._path, "r", encoding="utf-8") as read_file:
+                            existing_data = json.load(read_file)
+                            existing_data.append(log_entry)
+                    except (json.JSONDecodeError, FileNotFoundError):
+                        # If no valid JSON or file doesn't exist, start with an empty list
+                        existing_data = [log_entry]
+                    
+                    with open(self._path, "w", encoding="utf-8") as write_file:
+                        json.dump(existing_data, write_file, indent=4)
+                        write_file.write("\n")
+            
+            else:
+                # Append log in plain text format
+                message = f"{now}: " + ", ".join([f"{key}=\"{value}\"" for key, value in kwargs.items()]) + "\n"
+                with open(self._path, "a", encoding="utf-8") as file:
+                    file.write(message)
 
+        except Exception as e:
+            raise ValueError(f"Failed to log message: {str(e)}")
+        
 class PickleHandler:
     def __init__(self, file_name: str) -> None:
         """

src/crewai/utilities/internal_instructor.py

@@ -11,12 +11,10 @@ class InternalInstructor:
         model: Type,
         agent: Optional[Any] = None,
         llm: Optional[str] = None,
-        instructions: Optional[str] = None,
     ):
         self.content = content
         self.agent = agent
         self.llm = llm
-        self.instructions = instructions
         self.model = model
         self._client = None
         self.set_instructor()
@@ -31,10 +29,7 @@ class InternalInstructor:
             import instructor
             from litellm import completion
 
-            self._client = instructor.from_litellm(
-                completion,
-                mode=instructor.Mode.TOOLS,
-            )
+            self._client = instructor.from_litellm(completion)
 
     def to_json(self):
         model = self.to_pydantic()
@@ -42,8 +37,6 @@ class InternalInstructor:
 
     def to_pydantic(self):
         messages = [{"role": "user", "content": self.content}]
-        if self.instructions:
-            messages.append({"role": "system", "content": self.instructions})
         model = self._client.chat.completions.create(
             model=self.llm.model, response_model=self.model, messages=messages
         )

src/crewai/utilities/llm_utils.py

@@ -53,6 +53,7 @@ def create_llm(
         timeout: Optional[float] = getattr(llm_value, "timeout", None)
         api_key: Optional[str] = getattr(llm_value, "api_key", None)
         base_url: Optional[str] = getattr(llm_value, "base_url", None)
+        api_base: Optional[str] = getattr(llm_value, "api_base", None)
 
         created_llm = LLM(
             model=model,
@@ -62,6 +63,7 @@ def create_llm(
             timeout=timeout,
             api_key=api_key,
             base_url=base_url,
+            api_base=api_base,
         )
         return created_llm
     except Exception as e:
@@ -101,8 +103,18 @@ def _llm_via_environment_or_fallback() -> Optional[LLM]:
     callbacks: List[Any] = []
 
     # Optional base URL from env
-    api_base = os.environ.get("OPENAI_API_BASE") or os.environ.get("OPENAI_BASE_URL")
-    if api_base:
+    base_url = (
+        os.environ.get("BASE_URL")
+        or os.environ.get("OPENAI_API_BASE")
+        or os.environ.get("OPENAI_BASE_URL")
+    )
+
+    api_base = os.environ.get("API_BASE") or os.environ.get("AZURE_API_BASE")
+
+    # Synchronize base_url and api_base if one is populated and the other is not
+    if base_url and not api_base:
+        api_base = base_url
+    elif api_base and not base_url:
         base_url = api_base
 
     # Initialize llm_params dictionary
@@ -115,6 +127,7 @@ def _llm_via_environment_or_fallback() -> Optional[LLM]:
         "timeout": timeout,
         "api_key": api_key,
         "base_url": base_url,
+        "api_base": api_base,
         "api_version": api_version,
         "presence_penalty": presence_penalty,
         "frequency_penalty": frequency_penalty,

src/crewai/utilities/paths.py

@@ -5,14 +5,18 @@ import appdirs
 
 """Path management utilities for CrewAI storage and configuration."""
 
-def db_storage_path():
-    """Returns the path for database storage."""
+def db_storage_path() -> str:
+    """Returns the path for SQLite database storage.
+
+    Returns:
+        str: Full path to the SQLite database file
+    """
     app_name = get_project_directory_name()
     app_author = "CrewAI"
 
     data_dir = Path(appdirs.user_data_dir(app_name, app_author))
     data_dir.mkdir(parents=True, exist_ok=True)
-    return data_dir
+    return str(data_dir)
 
 
 def get_project_directory_name():
@@ -24,4 +28,4 @@ def get_project_directory_name():
     else:
         cwd = Path.cwd()
         project_directory_name = cwd.name
-        return project_directory_name
+        return project_directory_name

src/crewai/utilities/printer.py

@@ -21,6 +21,16 @@ class Printer:
             self._print_yellow(content)
         elif color == "bold_yellow":
             self._print_bold_yellow(content)
+        elif color == "cyan":
+            self._print_cyan(content)
+        elif color == "bold_cyan":
+            self._print_bold_cyan(content)
+        elif color == "magenta":
+            self._print_magenta(content)
+        elif color == "bold_magenta":
+            self._print_bold_magenta(content)
+        elif color == "green":
+            self._print_green(content)
         else:
             print(content)
 
@@ -44,3 +54,18 @@ class Printer:
 
     def _print_bold_yellow(self, content):
         print("\033[1m\033[93m {}\033[00m".format(content))
+
+    def _print_cyan(self, content):
+        print("\033[96m {}\033[00m".format(content))
+
+    def _print_bold_cyan(self, content):
+        print("\033[1m\033[96m {}\033[00m".format(content))
+
+    def _print_magenta(self, content):
+        print("\033[35m {}\033[00m".format(content))
+
+    def _print_bold_magenta(self, content):
+        print("\033[1m\033[35m {}\033[00m".format(content))
+
+    def _print_green(self, content):
+        print("\033[32m {}\033[00m".format(content))

src/crewai/utilities/pydantic_schema_parser.py

@@ -1,4 +1,4 @@
-from typing import Type, Union, get_args, get_origin
+from typing import Dict, List, Type, Union, get_args, get_origin
 
 from pydantic import BaseModel
 
@@ -10,40 +10,83 @@ class PydanticSchemaParser(BaseModel):
         """
         Public method to get the schema of a Pydantic model.
 
-        :param model: The Pydantic model class to generate schema for.
         :return: String representation of the model schema.
         """
-        return self._get_model_schema(self.model)
+        return "{\n" + self._get_model_schema(self.model) + "\n}"
 
-    def _get_model_schema(self, model, depth=0) -> str:
-        indent = "    " * depth
-        lines = [f"{indent}{{"]
-        for field_name, field in model.model_fields.items():
-            field_type_str = self._get_field_type(field, depth + 1)
-            lines.append(f"{indent}    {field_name}: {field_type_str},")
-        lines[-1] = lines[-1].rstrip(",")  # Remove trailing comma from last item
-        lines.append(f"{indent}}}")
-        return "\n".join(lines)
+    def _get_model_schema(self, model: Type[BaseModel], depth: int = 0) -> str:
+        indent = " " * 4 * depth
+        lines = [
+            f"{indent}    {field_name}: {self._get_field_type(field, depth + 1)}"
+            for field_name, field in model.model_fields.items()
+        ]
+        return ",\n".join(lines)
 
-    def _get_field_type(self, field, depth) -> str:
+    def _get_field_type(self, field, depth: int) -> str:
         field_type = field.annotation
-        if get_origin(field_type) is list:
+        origin = get_origin(field_type)
+
+        if origin in {list, List}:
             list_item_type = get_args(field_type)[0]
-            if isinstance(list_item_type, type) and issubclass(
-                list_item_type, BaseModel
-            ):
-                nested_schema = self._get_model_schema(list_item_type, depth + 1)
-                return f"List[\n{nested_schema}\n{' ' * 4 * depth}]"
+            return self._format_list_type(list_item_type, depth)
+
+        if origin in {dict, Dict}:
+            key_type, value_type = get_args(field_type)
+            return f"Dict[{key_type.__name__}, {value_type.__name__}]"
+
+        if origin is Union:
+            return self._format_union_type(field_type, depth)
+
+        if isinstance(field_type, type) and issubclass(field_type, BaseModel):
+            nested_schema = self._get_model_schema(field_type, depth)
+            nested_indent = " " * 4 * depth
+            return f"{field_type.__name__}\n{nested_indent}{{\n{nested_schema}\n{nested_indent}}}"
+
+        return field_type.__name__
+
+    def _format_list_type(self, list_item_type, depth: int) -> str:
+        if isinstance(list_item_type, type) and issubclass(list_item_type, BaseModel):
+            nested_schema = self._get_model_schema(list_item_type, depth + 1)
+            nested_indent = " " * 4 * (depth)
+            return f"List[\n{nested_indent}{{\n{nested_schema}\n{nested_indent}}}\n{nested_indent}]"
+        return f"List[{list_item_type.__name__}]"
+
+    def _format_union_type(self, field_type, depth: int) -> str:
+        args = get_args(field_type)
+        if type(None) in args:
+            # It's an Optional type
+            non_none_args = [arg for arg in args if arg is not type(None)]
+            if len(non_none_args) == 1:
+                inner_type = self._get_field_type_for_annotation(
+                    non_none_args[0], depth
+                )
+                return f"Optional[{inner_type}]"
             else:
-                return f"List[{list_item_type.__name__}]"
-        elif get_origin(field_type) is Union:
-            union_args = get_args(field_type)
-            if type(None) in union_args:
-                non_none_type = next(arg for arg in union_args if arg is not type(None))
-                return f"Optional[{self._get_field_type(field.__class__(annotation=non_none_type), depth)}]"
-            else:
-                return f"Union[{', '.join(arg.__name__ for arg in union_args)}]"
-        elif isinstance(field_type, type) and issubclass(field_type, BaseModel):
-            return self._get_model_schema(field_type, depth)
+                # Union with None and multiple other types
+                inner_types = ", ".join(
+                    self._get_field_type_for_annotation(arg, depth)
+                    for arg in non_none_args
+                )
+                return f"Optional[Union[{inner_types}]]"
         else:
-            return getattr(field_type, "__name__", str(field_type))
+            # General Union type
+            inner_types = ", ".join(
+                self._get_field_type_for_annotation(arg, depth) for arg in args
+            )
+            return f"Union[{inner_types}]"
+
+    def _get_field_type_for_annotation(self, annotation, depth: int) -> str:
+        origin = get_origin(annotation)
+        if origin in {list, List}:
+            list_item_type = get_args(annotation)[0]
+            return self._format_list_type(list_item_type, depth)
+        if origin in {dict, Dict}:
+            key_type, value_type = get_args(annotation)
+            return f"Dict[{key_type.__name__}, {value_type.__name__}]"
+        if origin is Union:
+            return self._format_union_type(annotation, depth)
+        if isinstance(annotation, type) and issubclass(annotation, BaseModel):
+            nested_schema = self._get_model_schema(annotation, depth)
+            nested_indent = " " * 4 * depth
+            return f"{annotation.__name__}\n{nested_indent}{{\n{nested_schema}\n{nested_indent}}}"
+        return annotation.__name__

src/crewai/utilities/token_counter_callback.py

@@ -23,11 +23,15 @@ class TokenCalcHandler(CustomLogger):
 
         with warnings.catch_warnings():
             warnings.simplefilter("ignore", UserWarning)
-            usage: Usage = response_obj["usage"]
-            self.token_cost_process.sum_successful_requests(1)
-            self.token_cost_process.sum_prompt_tokens(usage.prompt_tokens)
-            self.token_cost_process.sum_completion_tokens(usage.completion_tokens)
-            if usage.prompt_tokens_details:
-                self.token_cost_process.sum_cached_prompt_tokens(
-                    usage.prompt_tokens_details.cached_tokens
-                )
+            if isinstance(response_obj, dict) and "usage" in response_obj:
+                usage: Usage = response_obj["usage"]
+                if usage:
+                    self.token_cost_process.sum_successful_requests(1)
+                    if hasattr(usage, "prompt_tokens"):
+                        self.token_cost_process.sum_prompt_tokens(usage.prompt_tokens)
+                    if hasattr(usage, "completion_tokens"):
+                        self.token_cost_process.sum_completion_tokens(usage.completion_tokens)
+                    if hasattr(usage, "prompt_tokens_details") and usage.prompt_tokens_details:
+                        self.token_cost_process.sum_cached_prompt_tokens(
+                            usage.prompt_tokens_details.cached_tokens
+                        )

src/crewai/utilities/training_handler.py

@@ -1,3 +1,5 @@
+import os
+
 from crewai.utilities.file_handler import PickleHandler
 
 
@@ -29,3 +31,8 @@ class CrewTrainingHandler(PickleHandler):
             data[agent_id] = {train_iteration: new_data}
 
         self.save(data)
+
+    def clear(self) -> None:
+        """Clear the training data by removing the file or resetting its contents."""
+        if os.path.exists(self.file_path):
+            self.save({})

tests/agent_test.py

@@ -10,6 +10,7 @@ from crewai import Agent, Crew, Task
 from crewai.agents.cache import CacheHandler
 from crewai.agents.crew_agent_executor import CrewAgentExecutor
 from crewai.agents.parser import AgentAction, CrewAgentParser, OutputParserException
+from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
 from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
 from crewai.llm import LLM
 from crewai.tools import tool
@@ -114,35 +115,6 @@ def test_custom_llm_temperature_preservation():
     assert agent.llm.temperature == 0.7
 
 
-@pytest.mark.vcr(filter_headers=["authorization"])
-def test_agent_execute_task():
-    from langchain_openai import ChatOpenAI
-
-    from crewai import Task
-
-    agent = Agent(
-        role="Math Tutor",
-        goal="Solve math problems accurately",
-        backstory="You are an experienced math tutor with a knack for explaining complex concepts simply.",
-        llm=ChatOpenAI(temperature=0.7, model="gpt-4o-mini"),
-    )
-
-    task = Task(
-        description="Calculate the area of a circle with radius 5 cm.",
-        expected_output="The calculated area of the circle in square centimeters.",
-        agent=agent,
-    )
-
-    result = agent.execute_task(task)
-
-    assert result is not None
-    assert (
-        result
-        == "The calculated area of the circle is approximately 78.5 square centimeters."
-    )
-    assert "square centimeters" in result.lower()
-
-
 @pytest.mark.vcr(filter_headers=["authorization"])
 def test_agent_execution():
     agent = Agent(
@@ -1211,7 +1183,7 @@ def test_agent_max_retry_limit():
             [
                 mock.call(
                     {
-                        "input": "Say the word: Hi\n\nThis is the expect criteria for your final answer: The word: Hi\nyou MUST return the actual complete content as the final answer, not a summary.",
+                        "input": "Say the word: Hi\n\nThis is the expected criteria for your final answer: The word: Hi\nyou MUST return the actual complete content as the final answer, not a summary.",
                         "tool_names": "",
                         "tools": "",
                         "ask_for_human_input": True,
@@ -1219,7 +1191,7 @@ def test_agent_max_retry_limit():
                 ),
                 mock.call(
                     {
-                        "input": "Say the word: Hi\n\nThis is the expect criteria for your final answer: The word: Hi\nyou MUST return the actual complete content as the final answer, not a summary.",
+                        "input": "Say the word: Hi\n\nThis is the expected criteria for your final answer: The word: Hi\nyou MUST return the actual complete content as the final answer, not a summary.",
                         "tool_names": "",
                         "tools": "",
                         "ask_for_human_input": True,
@@ -1629,3 +1601,181 @@ def test_agent_with_knowledge_sources():
 
         # Assert that the agent provides the correct information
         assert "red" in result.raw.lower()
+
+
+@pytest.mark.vcr(filter_headers=["authorization"])
+def test_agent_with_knowledge_sources_works_with_copy():
+    content = "Brandon's favorite color is red and he likes Mexican food."
+    string_source = StringKnowledgeSource(content=content)
+
+    with patch(
+        "crewai.knowledge.source.base_knowledge_source.BaseKnowledgeSource",
+        autospec=True,
+    ) as MockKnowledgeSource:
+        mock_knowledge_source_instance = MockKnowledgeSource.return_value
+        mock_knowledge_source_instance.__class__ = BaseKnowledgeSource
+        mock_knowledge_source_instance.sources = [string_source]
+
+        agent = Agent(
+            role="Information Agent",
+            goal="Provide information based on knowledge sources",
+            backstory="You have access to specific knowledge sources.",
+            llm=LLM(model="gpt-4o-mini"),
+            knowledge_sources=[string_source],
+        )
+
+        with patch(
+            "crewai.knowledge.storage.knowledge_storage.KnowledgeStorage"
+        ) as MockKnowledgeStorage:
+            mock_knowledge_storage = MockKnowledgeStorage.return_value
+            agent.knowledge_storage = mock_knowledge_storage
+
+            agent_copy = agent.copy()
+
+            assert agent_copy.role == agent.role
+            assert agent_copy.goal == agent.goal
+            assert agent_copy.backstory == agent.backstory
+            assert agent_copy.knowledge_sources is not None
+            assert len(agent_copy.knowledge_sources) == 1
+            assert isinstance(agent_copy.knowledge_sources[0], StringKnowledgeSource)
+            assert agent_copy.knowledge_sources[0].content == content
+            assert isinstance(agent_copy.llm, LLM)
+
+
+@pytest.mark.vcr(filter_headers=["authorization"])
+def test_litellm_auth_error_handling():
+    """Test that LiteLLM authentication errors are handled correctly and not retried."""
+    from litellm import AuthenticationError as LiteLLMAuthenticationError
+
+    # Create an agent with a mocked LLM and max_retry_limit=0
+    agent = Agent(
+        role="test role",
+        goal="test goal",
+        backstory="test backstory",
+        llm=LLM(model="gpt-4"),
+        max_retry_limit=0,  # Disable retries for authentication errors
+    )
+
+    # Create a task
+    task = Task(
+        description="Test task",
+        expected_output="Test output",
+        agent=agent,
+    )
+
+    # Mock the LLM call to raise AuthenticationError
+    with (
+        patch.object(LLM, "call") as mock_llm_call,
+        pytest.raises(LiteLLMAuthenticationError, match="Invalid API key"),
+    ):
+        mock_llm_call.side_effect = LiteLLMAuthenticationError(
+            message="Invalid API key", llm_provider="openai", model="gpt-4"
+        )
+        agent.execute_task(task)
+
+    # Verify the call was only made once (no retries)
+    mock_llm_call.assert_called_once()
+
+
+def test_crew_agent_executor_litellm_auth_error():
+    """Test that CrewAgentExecutor handles LiteLLM authentication errors by raising them."""
+    from litellm.exceptions import AuthenticationError
+
+    from crewai.agents.tools_handler import ToolsHandler
+    from crewai.utilities import Printer
+
+    # Create an agent and executor
+    agent = Agent(
+        role="test role",
+        goal="test goal",
+        backstory="test backstory",
+        llm=LLM(model="gpt-4", api_key="invalid_api_key"),
+    )
+    task = Task(
+        description="Test task",
+        expected_output="Test output",
+        agent=agent,
+    )
+
+    # Create executor with all required parameters
+    executor = CrewAgentExecutor(
+        agent=agent,
+        task=task,
+        llm=agent.llm,
+        crew=None,
+        prompt={"system": "You are a test agent", "user": "Execute the task: {input}"},
+        max_iter=5,
+        tools=[],
+        tools_names="",
+        stop_words=[],
+        tools_description="",
+        tools_handler=ToolsHandler(),
+    )
+
+    # Mock the LLM call to raise AuthenticationError
+    with (
+        patch.object(LLM, "call") as mock_llm_call,
+        patch.object(Printer, "print") as mock_printer,
+        pytest.raises(AuthenticationError) as exc_info,
+    ):
+        mock_llm_call.side_effect = AuthenticationError(
+            message="Invalid API key", llm_provider="openai", model="gpt-4"
+        )
+        executor.invoke(
+            {
+                "input": "test input",
+                "tool_names": "",
+                "tools": "",
+            }
+        )
+
+    # Verify error handling messages
+    error_message = f"Error during LLM call: {str(mock_llm_call.side_effect)}"
+    mock_printer.assert_any_call(
+        content=error_message,
+        color="red",
+    )
+
+    # Verify the call was only made once (no retries)
+    mock_llm_call.assert_called_once()
+
+    # Assert that the exception was raised and has the expected attributes
+    assert exc_info.type is AuthenticationError
+    assert "Invalid API key".lower() in exc_info.value.message.lower()
+    assert exc_info.value.llm_provider == "openai"
+    assert exc_info.value.model == "gpt-4"
+
+
+def test_litellm_anthropic_error_handling():
+    """Test that AnthropicError from LiteLLM is handled correctly and not retried."""
+    from litellm.llms.anthropic.common_utils import AnthropicError
+
+    # Create an agent with a mocked LLM that uses an Anthropic model
+    agent = Agent(
+        role="test role",
+        goal="test goal",
+        backstory="test backstory",
+        llm=LLM(model="claude-3.5-sonnet-20240620"),
+        max_retry_limit=0,
+    )
+
+    # Create a task
+    task = Task(
+        description="Test task",
+        expected_output="Test output",
+        agent=agent,
+    )
+
+    # Mock the LLM call to raise AnthropicError
+    with (
+        patch.object(LLM, "call") as mock_llm_call,
+        pytest.raises(AnthropicError, match="Test Anthropic error"),
+    ):
+        mock_llm_call.side_effect = AnthropicError(
+            status_code=500,
+            message="Test Anthropic error",
+        )
+        agent.execute_task(task)
+
+    # Verify the LLM call was only made once (no retries)
+    mock_llm_call.assert_called_once()

tests/cassettes/test_agent_error_on_parsing_tool.yaml

@@ -2,21 +2,21 @@ interactions:
 - request:
     body: '{"messages": [{"role": "system", "content": "You are test role. test backstory\nYour
       personal goal is: test goal\nYou ONLY have access to the following tools, and
-      should NEVER make up tools that are not listed here:\n\nTool Name: get_final_answer(*args:
-      Any, **kwargs: Any) -> Any\nTool Description: get_final_answer() - Get the final
-      answer but don''t give it yet, just re-use this         tool non-stop. \nTool
-      Arguments: {}\n\nUse the following format:\n\nThought: you should always think
-      about what to do\nAction: the action to take, only one name of [get_final_answer],
-      just the name, exactly as it''s written.\nAction Input: the input to the action,
-      just a simple python dictionary, enclosed in curly braces, using \" to wrap
-      keys and values.\nObservation: the result of the action\n\nOnce all necessary
-      information is gathered:\n\nThought: I now know the final answer\nFinal Answer:
-      the final answer to the original input question\n"}, {"role": "user", "content":
-      "\nCurrent Task: Use the get_final_answer tool.\n\nThis is the expect criteria
-      for your final answer: The final answer\nyou MUST return the actual complete
-      content as the final answer, not a summary.\n\nBegin! This is VERY important
-      to you, use the tools available and give your best Final Answer, your job depends
-      on it!\n\nThought:"}], "model": "gpt-4o"}'
+      should NEVER make up tools that are not listed here:\n\nTool Name: get_final_answer\nTool
+      Arguments: {}\nTool Description: Get the final answer but don''t give it yet,
+      just re-use this\n        tool non-stop.\n\nIMPORTANT: Use the following format
+      in your response:\n\n```\nThought: you should always think about what to do\nAction:
+      the action to take, only one name of [get_final_answer], just the name, exactly
+      as it''s written.\nAction Input: the input to the action, just a simple JSON
+      object, enclosed in curly braces, using \" to wrap keys and values.\nObservation:
+      the result of the action\n```\n\nOnce all necessary information is gathered,
+      return the following format:\n\n```\nThought: I now know the final answer\nFinal
+      Answer: the final answer to the original input question\n```"}, {"role": "user",
+      "content": "\nCurrent Task: Use the get_final_answer tool.\n\nThis is the expect
+      criteria for your final answer: The final answer\nyou MUST return the actual
+      complete content as the final answer, not a summary.\n\nBegin! This is VERY
+      important to you, use the tools available and give your best Final Answer, your
+      job depends on it!\n\nThought:"}], "model": "gpt-4o", "stop": ["\nObservation:"]}'
     headers:
       accept:
       - application/json
@@ -25,16 +25,13 @@ interactions:
       connection:
       - keep-alive
       content-length:
-      - '1325'
+      - '1367'
       content-type:
       - application/json
-      cookie:
-      - _cfuvid=ePJSDFdHag2D8lj21_ijAMWjoA6xfnPNxN4uekvC728-1727226247743-0.0.1.1-604800000;
-        __cf_bm=3giyBOIM0GNudFELtsBWYXwLrpLBTNLsh81wfXgu2tg-1727226247-1.0.1.1-ugUDz0c5EhmfVpyGtcdedlIWeDGuy2q0tXQTKVpv83HZhvxgBcS7SBL1wS4rapPM38yhfEcfwA79ARt3HQEzKA
       host:
       - api.openai.com
       user-agent:
-      - OpenAI/Python 1.47.0
+      - OpenAI/Python 1.59.6
       x-stainless-arch:
       - arm64
       x-stainless-async:
@@ -44,30 +41,35 @@ interactions:
       x-stainless-os:
       - MacOS
       x-stainless-package-version:
-      - 1.47.0
+      - 1.59.6
       x-stainless-raw-response:
       - 'true'
+      x-stainless-retry-count:
+      - '0'
       x-stainless-runtime:
       - CPython
       x-stainless-runtime-version:
-      - 3.11.7
+      - 3.12.7
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-ABAtOWmVjvzQ9X58tKAUcOF4gmXwx\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1727226842,\n  \"model\": \"gpt-4o-2024-05-13\",\n
+    content: "{\n  \"id\": \"chatcmpl-AsXdf4OZKCZSigmN4k0gyh67NciqP\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1737562383,\n  \"model\": \"gpt-4o-2024-08-06\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"Thought: I need to use the get_final_answer
-      tool to determine the final answer.\\nAction: get_final_answer\\nAction Input:
-      {}\",\n        \"refusal\": null\n      },\n      \"logprobs\": null,\n      \"finish_reason\":
-      \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 274,\n    \"completion_tokens\":
-      27,\n    \"total_tokens\": 301,\n    \"completion_tokens_details\": {\n      \"reasoning_tokens\":
-      0\n    }\n  },\n  \"system_fingerprint\": \"fp_e375328146\"\n}\n"
+      \"assistant\",\n        \"content\": \"```\\nThought: I have to use the available
+      tool to get the final answer. Let's proceed with executing it.\\nAction: get_final_answer\\nAction
+      Input: {}\",\n        \"refusal\": null\n      },\n      \"logprobs\": null,\n
+      \     \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
+      274,\n    \"completion_tokens\": 33,\n    \"total_tokens\": 307,\n    \"prompt_tokens_details\":
+      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
+      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"service_tier\":
+      \"default\",\n  \"system_fingerprint\": \"fp_50cad350e4\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8c8727b3492f31e6-MIA
+      - 9060d43e3be1d690-IAD
       Connection:
       - keep-alive
       Content-Encoding:
@@ -75,19 +77,27 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 25 Sep 2024 01:14:03 GMT
+      - Wed, 22 Jan 2025 16:13:03 GMT
       Server:
       - cloudflare
+      Set-Cookie:
+      - __cf_bm=_Jcp7wnO_mXdvOnborCN6j8HwJxJXbszedJC1l7pFUg-1737562383-1.0.1.1-pDSLXlg.nKjG4wsT7mTJPjUvOX1UJITiS4MqKp6yfMWwRSJINsW1qC48SAcjBjakx2H5I1ESVk9JtUpUFDtf4g;
+        path=/; expires=Wed, 22-Jan-25 16:43:03 GMT; domain=.api.openai.com; HttpOnly;
+        Secure; SameSite=None
+      - _cfuvid=x3SYvzL2nq_PTBGtE8R9cl5CkeaaDzZFQIrYfo91S2s-1737562383916-0.0.1.1-604800000;
+        path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
       Transfer-Encoding:
       - chunked
       X-Content-Type-Options:
       - nosniff
       access-control-expose-headers:
       - X-Request-ID
+      alt-svc:
+      - h3=":443"; ma=86400
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '348'
+      - '791'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -99,45 +109,59 @@ interactions:
       x-ratelimit-remaining-requests:
       - '9999'
       x-ratelimit-remaining-tokens:
-      - '29999682'
+      - '29999680'
       x-ratelimit-reset-requests:
       - 6ms
       x-ratelimit-reset-tokens:
       - 0s
       x-request-id:
-      - req_be929caac49706f487950548bdcdd46e
+      - req_eeed99acafd3aeb1e3d4a6c8063192b0
     http_version: HTTP/1.1
     status_code: 200
 - request:
     body: '{"messages": [{"role": "system", "content": "You are test role. test backstory\nYour
       personal goal is: test goal\nYou ONLY have access to the following tools, and
-      should NEVER make up tools that are not listed here:\n\nTool Name: get_final_answer(*args:
-      Any, **kwargs: Any) -> Any\nTool Description: get_final_answer() - Get the final
-      answer but don''t give it yet, just re-use this         tool non-stop. \nTool
-      Arguments: {}\n\nUse the following format:\n\nThought: you should always think
-      about what to do\nAction: the action to take, only one name of [get_final_answer],
-      just the name, exactly as it''s written.\nAction Input: the input to the action,
-      just a simple python dictionary, enclosed in curly braces, using \" to wrap
-      keys and values.\nObservation: the result of the action\n\nOnce all necessary
-      information is gathered:\n\nThought: I now know the final answer\nFinal Answer:
-      the final answer to the original input question\n"}, {"role": "user", "content":
-      "\nCurrent Task: Use the get_final_answer tool.\n\nThis is the expect criteria
-      for your final answer: The final answer\nyou MUST return the actual complete
-      content as the final answer, not a summary.\n\nBegin! This is VERY important
-      to you, use the tools available and give your best Final Answer, your job depends
-      on it!\n\nThought:"}, {"role": "user", "content": "Thought: I need to use the
-      get_final_answer tool to determine the final answer.\nAction: get_final_answer\nAction
+      should NEVER make up tools that are not listed here:\n\nTool Name: get_final_answer\nTool
+      Arguments: {}\nTool Description: Get the final answer but don''t give it yet,
+      just re-use this\n        tool non-stop.\n\nIMPORTANT: Use the following format
+      in your response:\n\n```\nThought: you should always think about what to do\nAction:
+      the action to take, only one name of [get_final_answer], just the name, exactly
+      as it''s written.\nAction Input: the input to the action, just a simple JSON
+      object, enclosed in curly braces, using \" to wrap keys and values.\nObservation:
+      the result of the action\n```\n\nOnce all necessary information is gathered,
+      return the following format:\n\n```\nThought: I now know the final answer\nFinal
+      Answer: the final answer to the original input question\n```"}, {"role": "user",
+      "content": "\nCurrent Task: Use the get_final_answer tool.\n\nThis is the expect
+      criteria for your final answer: The final answer\nyou MUST return the actual
+      complete content as the final answer, not a summary.\n\nBegin! This is VERY
+      important to you, use the tools available and give your best Final Answer, your
+      job depends on it!\n\nThought:"}, {"role": "assistant", "content": "```\nThought:
+      I have to use the available tool to get the final answer. Let''s proceed with
+      executing it.\nAction: get_final_answer\nAction Input: {}\nObservation: I encountered
+      an error: Error on parsing tool.\nMoving on then. I MUST either use a tool (use
+      one at time) OR give my best final answer not both at the same time. When responding,
+      I must use the following format:\n\n```\nThought: you should always think about
+      what to do\nAction: the action to take, should be one of [get_final_answer]\nAction
+      Input: the input to the action, dictionary enclosed in curly braces\nObservation:
+      the result of the action\n```\nThis Thought/Action/Action Input/Result can repeat
+      N times. Once I know the final answer, I must return the following format:\n\n```\nThought:
+      I now can give a great answer\nFinal Answer: Your final answer must be the great
+      and the most complete as possible, it must be outcome described\n\n```"}, {"role":
+      "assistant", "content": "```\nThought: I have to use the available tool to get
+      the final answer. Let''s proceed with executing it.\nAction: get_final_answer\nAction
       Input: {}\nObservation: I encountered an error: Error on parsing tool.\nMoving
       on then. I MUST either use a tool (use one at time) OR give my best final answer
-      not both at the same time. To Use the following format:\n\nThought: you should
-      always think about what to do\nAction: the action to take, should be one of
-      [get_final_answer]\nAction Input: the input to the action, dictionary enclosed
-      in curly braces\nObservation: the result of the action\n... (this Thought/Action/Action
-      Input/Result can repeat N times)\nThought: I now can give a great answer\nFinal
+      not both at the same time. When responding, I must use the following format:\n\n```\nThought:
+      you should always think about what to do\nAction: the action to take, should
+      be one of [get_final_answer]\nAction Input: the input to the action, dictionary
+      enclosed in curly braces\nObservation: the result of the action\n```\nThis Thought/Action/Action
+      Input/Result can repeat N times. Once I know the final answer, I must return
+      the following format:\n\n```\nThought: I now can give a great answer\nFinal
       Answer: Your final answer must be the great and the most complete as possible,
-      it must be outcome described\n\n \nNow it''s time you MUST give your absolute
+      it must be outcome described\n\n```\nNow it''s time you MUST give your absolute
       best final answer. You''ll ignore all previous instructions, stop using any
-      tools, and just return your absolute BEST Final answer."}], "model": "gpt-4o"}'
+      tools, and just return your absolute BEST Final answer."}], "model": "gpt-4o",
+      "stop": ["\nObservation:"]}'
     headers:
       accept:
       - application/json
@@ -146,16 +170,16 @@ interactions:
       connection:
       - keep-alive
       content-length:
-      - '2320'
+      - '3445'
       content-type:
       - application/json
       cookie:
-      - _cfuvid=ePJSDFdHag2D8lj21_ijAMWjoA6xfnPNxN4uekvC728-1727226247743-0.0.1.1-604800000;
-        __cf_bm=3giyBOIM0GNudFELtsBWYXwLrpLBTNLsh81wfXgu2tg-1727226247-1.0.1.1-ugUDz0c5EhmfVpyGtcdedlIWeDGuy2q0tXQTKVpv83HZhvxgBcS7SBL1wS4rapPM38yhfEcfwA79ARt3HQEzKA
+      - __cf_bm=_Jcp7wnO_mXdvOnborCN6j8HwJxJXbszedJC1l7pFUg-1737562383-1.0.1.1-pDSLXlg.nKjG4wsT7mTJPjUvOX1UJITiS4MqKp6yfMWwRSJINsW1qC48SAcjBjakx2H5I1ESVk9JtUpUFDtf4g;
+        _cfuvid=x3SYvzL2nq_PTBGtE8R9cl5CkeaaDzZFQIrYfo91S2s-1737562383916-0.0.1.1-604800000
       host:
       - api.openai.com
       user-agent:
-      - OpenAI/Python 1.47.0
+      - OpenAI/Python 1.59.6
       x-stainless-arch:
       - arm64
       x-stainless-async:
@@ -165,29 +189,36 @@ interactions:
       x-stainless-os:
       - MacOS
       x-stainless-package-version:
-      - 1.47.0
+      - 1.59.6
       x-stainless-raw-response:
       - 'true'
+      x-stainless-retry-count:
+      - '0'
       x-stainless-runtime:
       - CPython
       x-stainless-runtime-version:
-      - 3.11.7
+      - 3.12.7
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-ABAtPaaeRfdNsZ3k06CfAmrEW8IJu\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1727226843,\n  \"model\": \"gpt-4o-2024-05-13\",\n
+    content: "{\n  \"id\": \"chatcmpl-AsXdg9UrLvAiqWP979E6DszLsQ84k\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1737562384,\n  \"model\": \"gpt-4o-2024-08-06\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"Final Answer: The final answer\",\n        \"refusal\":
-      null\n      },\n      \"logprobs\": null,\n      \"finish_reason\": \"stop\"\n
-      \   }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 483,\n    \"completion_tokens\":
-      6,\n    \"total_tokens\": 489,\n    \"completion_tokens_details\": {\n      \"reasoning_tokens\":
-      0\n    }\n  },\n  \"system_fingerprint\": \"fp_e375328146\"\n}\n"
+      \"assistant\",\n        \"content\": \"```\\nThought: I now know the final answer\\nFinal
+      Answer: The final answer must be the great and the most complete as possible,
+      it must be outcome described.\\n```\",\n        \"refusal\": null\n      },\n
+      \     \"logprobs\": null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n
+      \ \"usage\": {\n    \"prompt_tokens\": 719,\n    \"completion_tokens\": 35,\n
+      \   \"total_tokens\": 754,\n    \"prompt_tokens_details\": {\n      \"cached_tokens\":
+      0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
+      \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"service_tier\":
+      \"default\",\n  \"system_fingerprint\": \"fp_50cad350e4\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8c8727b9da1f31e6-MIA
+      - 9060d4441edad690-IAD
       Connection:
       - keep-alive
       Content-Encoding:
@@ -195,7 +226,7 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 25 Sep 2024 01:14:03 GMT
+      - Wed, 22 Jan 2025 16:13:05 GMT
       Server:
       - cloudflare
       Transfer-Encoding:
@@ -209,7 +240,7 @@ interactions:
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '188'
+      - '928'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -221,13 +252,13 @@ interactions:
       x-ratelimit-remaining-requests:
       - '9999'
       x-ratelimit-remaining-tokens:
-      - '29999445'
+      - '29999187'
       x-ratelimit-reset-requests:
       - 6ms
       x-ratelimit-reset-tokens:
       - 1ms
       x-request-id:
-      - req_d8e32538689fe064627468bad802d9a8
+      - req_61fc7506e6db326ec572224aec81ef23
     http_version: HTTP/1.1
     status_code: 200
 version: 1

tests/cassettes/test_agent_execute_task.yaml

@@ -1,121 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Math Tutor. You are
-      an experienced math tutor with a knack for explaining complex concepts simply.\nYour
-      personal goal is: Solve math problems accurately\nTo give my best complete final
-      answer to the task use the exact following format:\n\nThought: I now can give
-      a great answer\nFinal Answer: Your final answer must be the great and the most
-      complete as possible, it must be outcome described.\n\nI MUST use these formats,
-      my job depends on it!"}, {"role": "user", "content": "\nCurrent Task: Calculate
-      the area of a circle with radius 5 cm.\n\nThis is the expect criteria for your
-      final answer: The calculated area of the circle in square centimeters.\nyou
-      MUST return the actual complete content as the final answer, not a summary.\n\nBegin!
-      This is VERY important to you, use the tools available and give your best Final
-      Answer, your job depends on it!\n\nThought:"}], "model": "gpt-4o-mini", "temperature":
-      0.7}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '969'
-      content-type:
-      - application/json
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.47.0
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.47.0
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.11.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AB7LEfa5gX4cncpI4avsK0CJG8pCb\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1727213192,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer\\n\\nTo
-      calculate the area of a circle, we use the formula:\\n\\n\\\\[ A = \\\\pi r^2
-      \\\\]\\n\\nwhere \\\\( A \\\\) is the area, \\\\( \\\\pi \\\\) (approximately
-      3.14), and \\\\( r \\\\) is the radius of the circle.\\n\\nGiven that the radius
-      \\\\( r \\\\) is 5 cm, we can substitute this value into the formula:\\n\\n\\\\[
-      A = \\\\pi (5 \\\\, \\\\text{cm})^2 \\\\]\\n\\nCalculating this step-by-step:\\n\\n1.
-      First, square the radius:\\n   \\\\[ (5 \\\\, \\\\text{cm})^2 = 25 \\\\, \\\\text{cm}^2
-      \\\\]\\n\\n2. Then, multiply by \\\\( \\\\pi \\\\):\\n   \\\\[ A = \\\\pi \\\\times
-      25 \\\\, \\\\text{cm}^2 \\\\]\\n\\nUsing the approximate value of \\\\( \\\\pi
-      \\\\):\\n   \\\\[ A \\\\approx 3.14 \\\\times 25 \\\\, \\\\text{cm}^2 \\\\]\\n
-      \  \\\\[ A \\\\approx 78.5 \\\\, \\\\text{cm}^2 \\\\]\\n\\nThus, the area of
-      the circle is approximately 78.5 square centimeters.\\n\\nFinal Answer: The
-      calculated area of the circle is approximately 78.5 square centimeters.\",\n
-      \       \"refusal\": null\n      },\n      \"logprobs\": null,\n      \"finish_reason\":
-      \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 182,\n    \"completion_tokens\":
-      270,\n    \"total_tokens\": 452,\n    \"completion_tokens_details\": {\n      \"reasoning_tokens\":
-      0\n    }\n  },\n  \"system_fingerprint\": \"fp_1bb46167f9\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8c85da71fcac1cf3-GRU
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Tue, 24 Sep 2024 21:26:34 GMT
-      Server:
-      - cloudflare
-      Set-Cookie:
-      - __cf_bm=rb61BZH2ejzD5YPmLaEJqI7km71QqyNJGTVdNxBq6qk-1727213194-1.0.1.1-pJ49onmgX9IugEMuYQMralzD7oj_6W.CHbSu4Su1z3NyjTGYg.rhgJZWng8feFYah._oSnoYlkTjpK1Wd2C9FA;
-        path=/; expires=Tue, 24-Sep-24 21:56:34 GMT; domain=.api.openai.com; HttpOnly;
-        Secure; SameSite=None
-      - _cfuvid=lbRdAddVWV6W3f5Dm9SaOPWDUOxqtZBSPr_fTW26nEA-1727213194587-0.0.1.1-604800000;
-        path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '2244'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '30000'
-      x-ratelimit-limit-tokens:
-      - '150000000'
-      x-ratelimit-remaining-requests:
-      - '29999'
-      x-ratelimit-remaining-tokens:
-      - '149999774'
-      x-ratelimit-reset-requests:
-      - 2ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_2e565b5f24c38968e4e923a47ecc6233
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_agent_human_input.yaml

@@ -1,4 +1,87 @@
 interactions:
+- request:
+    body: !!binary |
+      CqcXCiQKIgoMc2VydmljZS5uYW1lEhIKEGNyZXdBSS10ZWxlbWV0cnkS/hYKEgoQY3Jld2FpLnRl
+      bGVtZXRyeRJ5ChBuJJtOdNaB05mOW/p3915eEgj2tkAd3rZcASoQVG9vbCBVc2FnZSBFcnJvcjAB
+      OYa7/URvKBUYQUpcFEVvKBUYShoKDmNyZXdhaV92ZXJzaW9uEggKBjAuODYuMEoPCgNsbG0SCAoG
+      Z3B0LTRvegIYAYUBAAEAABLJBwoQifhX01E5i+5laGdALAlZBBIIBuGM1aN+OPgqDENyZXcgQ3Jl
+      YXRlZDABORVGruBvKBUYQaipwOBvKBUYShoKDmNyZXdhaV92ZXJzaW9uEggKBjAuODYuMEoaCg5w
+      eXRob25fdmVyc2lvbhIICgYzLjEyLjdKLgoIY3Jld19rZXkSIgogN2U2NjA4OTg5ODU5YTY3ZWVj
+      ODhlZWY3ZmNlODUyMjVKMQoHY3Jld19pZBImCiRiOThiNWEwMC01YTI1LTQxMDctYjQwNS1hYmYz
+      MjBhOGYzYThKHAoMY3Jld19wcm9jZXNzEgwKCnNlcXVlbnRpYWxKEQoLY3Jld19tZW1vcnkSAhAA
+      ShoKFGNyZXdfbnVtYmVyX29mX3Rhc2tzEgIYAUobChVjcmV3X251bWJlcl9vZl9hZ2VudHMSAhgB
+      SuQCCgtjcmV3X2FnZW50cxLUAgrRAlt7ImtleSI6ICIyMmFjZDYxMWU0NGVmNWZhYzA1YjUzM2Q3
+      NWU4ODkzYiIsICJpZCI6ICJkNWIyMzM1YS0yMmIyLTQyZWEtYmYwNS03OTc3NmU3MmYzOTIiLCAi
+      cm9sZSI6ICJEYXRhIFNjaWVudGlzdCIsICJ2ZXJib3NlPyI6IGZhbHNlLCAibWF4X2l0ZXIiOiAy
+      MCwgIm1heF9ycG0iOiBudWxsLCAiZnVuY3Rpb25fY2FsbGluZ19sbG0iOiAiIiwgImxsbSI6ICJn
+      cHQtNG8tbWluaSIsICJkZWxlZ2F0aW9uX2VuYWJsZWQ/IjogZmFsc2UsICJhbGxvd19jb2RlX2V4
+      ZWN1dGlvbj8iOiBmYWxzZSwgIm1heF9yZXRyeV9saW1pdCI6IDIsICJ0b29sc19uYW1lcyI6IFsi
+      Z2V0IGdyZWV0aW5ncyJdfV1KkgIKCmNyZXdfdGFza3MSgwIKgAJbeyJrZXkiOiAiYTI3N2IzNGIy
+      YzE0NmYwYzU2YzVlMTM1NmU4ZjhhNTciLCAiaWQiOiAiMjJiZWMyMzEtY2QyMS00YzU4LTgyN2Ut
+      MDU4MWE4ZjBjMTExIiwgImFzeW5jX2V4ZWN1dGlvbj8iOiBmYWxzZSwgImh1bWFuX2lucHV0PyI6
+      IGZhbHNlLCAiYWdlbnRfcm9sZSI6ICJEYXRhIFNjaWVudGlzdCIsICJhZ2VudF9rZXkiOiAiMjJh
+      Y2Q2MTFlNDRlZjVmYWMwNWI1MzNkNzVlODg5M2IiLCAidG9vbHNfbmFtZXMiOiBbImdldCBncmVl
+      dGluZ3MiXX1degIYAYUBAAEAABKOAgoQ5WYoxRtTyPjge4BduhL0rRIIv2U6rvWALfwqDFRhc2sg
+      Q3JlYXRlZDABOX068uBvKBUYQZkv8+BvKBUYSi4KCGNyZXdfa2V5EiIKIDdlNjYwODk4OTg1OWE2
+      N2VlYzg4ZWVmN2ZjZTg1MjI1SjEKB2NyZXdfaWQSJgokYjk4YjVhMDAtNWEyNS00MTA3LWI0MDUt
+      YWJmMzIwYThmM2E4Si4KCHRhc2tfa2V5EiIKIGEyNzdiMzRiMmMxNDZmMGM1NmM1ZTEzNTZlOGY4
+      YTU3SjEKB3Rhc2tfaWQSJgokMjJiZWMyMzEtY2QyMS00YzU4LTgyN2UtMDU4MWE4ZjBjMTExegIY
+      AYUBAAEAABKQAQoQXyeDtJDFnyp2Fjk9YEGTpxIIaNE7gbhPNYcqClRvb2wgVXNhZ2UwATkaXTvj
+      bygVGEGvx0rjbygVGEoaCg5jcmV3YWlfdmVyc2lvbhIICgYwLjg2LjBKHAoJdG9vbF9uYW1lEg8K
+      DUdldCBHcmVldGluZ3NKDgoIYXR0ZW1wdHMSAhgBegIYAYUBAAEAABLVBwoQMWfznt0qwauEzl7T
+      UOQxRBII9q+pUS5EdLAqDENyZXcgQ3JlYXRlZDABORONPORvKBUYQSAoS+RvKBUYShoKDmNyZXdh
+      aV92ZXJzaW9uEggKBjAuODYuMEoaCg5weXRob25fdmVyc2lvbhIICgYzLjEyLjdKLgoIY3Jld19r
+      ZXkSIgogYzMwNzYwMDkzMjY3NjE0NDRkNTdjNzFkMWRhM2YyN2NKMQoHY3Jld19pZBImCiQ3OTQw
+      MTkyNS1iOGU5LTQ3MDgtODUzMC00NDhhZmEzYmY4YjBKHAoMY3Jld19wcm9jZXNzEgwKCnNlcXVl
+      bnRpYWxKEQoLY3Jld19tZW1vcnkSAhAAShoKFGNyZXdfbnVtYmVyX29mX3Rhc2tzEgIYAUobChVj
+      cmV3X251bWJlcl9vZl9hZ2VudHMSAhgBSuoCCgtjcmV3X2FnZW50cxLaAgrXAlt7ImtleSI6ICI5
+      OGYzYjFkNDdjZTk2OWNmMDU3NzI3Yjc4NDE0MjVjZCIsICJpZCI6ICI5OTJkZjYyZi1kY2FiLTQy
+      OTUtOTIwNi05MDBkNDExNGIxZTkiLCAicm9sZSI6ICJGcmllbmRseSBOZWlnaGJvciIsICJ2ZXJi
+      b3NlPyI6IGZhbHNlLCAibWF4X2l0ZXIiOiAyMCwgIm1heF9ycG0iOiBudWxsLCAiZnVuY3Rpb25f
+      Y2FsbGluZ19sbG0iOiAiIiwgImxsbSI6ICJncHQtNG8tbWluaSIsICJkZWxlZ2F0aW9uX2VuYWJs
+      ZWQ/IjogZmFsc2UsICJhbGxvd19jb2RlX2V4ZWN1dGlvbj8iOiBmYWxzZSwgIm1heF9yZXRyeV9s
+      aW1pdCI6IDIsICJ0b29sc19uYW1lcyI6IFsiZGVjaWRlIGdyZWV0aW5ncyJdfV1KmAIKCmNyZXdf
+      dGFza3MSiQIKhgJbeyJrZXkiOiAiODBkN2JjZDQ5MDk5MjkwMDgzODMyZjBlOTgzMzgwZGYiLCAi
+      aWQiOiAiMmZmNjE5N2UtYmEyNy00YjczLWI0YTctNGZhMDQ4ZTYyYjQ3IiwgImFzeW5jX2V4ZWN1
+      dGlvbj8iOiBmYWxzZSwgImh1bWFuX2lucHV0PyI6IGZhbHNlLCAiYWdlbnRfcm9sZSI6ICJGcmll
+      bmRseSBOZWlnaGJvciIsICJhZ2VudF9rZXkiOiAiOThmM2IxZDQ3Y2U5NjljZjA1NzcyN2I3ODQx
+      NDI1Y2QiLCAidG9vbHNfbmFtZXMiOiBbImRlY2lkZSBncmVldGluZ3MiXX1degIYAYUBAAEAABKO
+      AgoQnjTp5boK7/+DQxztYIpqihIIgGnMUkBtzHEqDFRhc2sgQ3JlYXRlZDABOcpYcuRvKBUYQalE
+      c+RvKBUYSi4KCGNyZXdfa2V5EiIKIGMzMDc2MDA5MzI2NzYxNDQ0ZDU3YzcxZDFkYTNmMjdjSjEK
+      B2NyZXdfaWQSJgokNzk0MDE5MjUtYjhlOS00NzA4LTg1MzAtNDQ4YWZhM2JmOGIwSi4KCHRhc2tf
+      a2V5EiIKIDgwZDdiY2Q0OTA5OTI5MDA4MzgzMmYwZTk4MzM4MGRmSjEKB3Rhc2tfaWQSJgokMmZm
+      NjE5N2UtYmEyNy00YjczLWI0YTctNGZhMDQ4ZTYyYjQ3egIYAYUBAAEAABKTAQoQ26H9pLUgswDN
+      p9XhJwwL6BIIx3bw7mAvPYwqClRvb2wgVXNhZ2UwATmy7NPlbygVGEEvb+HlbygVGEoaCg5jcmV3
+      YWlfdmVyc2lvbhIICgYwLjg2LjBKHwoJdG9vbF9uYW1lEhIKEERlY2lkZSBHcmVldGluZ3NKDgoI
+      YXR0ZW1wdHMSAhgBegIYAYUBAAEAAA==
+    headers:
+      Accept:
+      - '*/*'
+      Accept-Encoding:
+      - gzip, deflate
+      Connection:
+      - keep-alive
+      Content-Length:
+      - '2986'
+      Content-Type:
+      - application/x-protobuf
+      User-Agent:
+      - OTel-OTLP-Exporter-Python/1.27.0
+    method: POST
+    uri: https://telemetry.crewai.com:4319/v1/traces
+  response:
+    body:
+      string: "\n\0"
+    headers:
+      Content-Length:
+      - '2'
+      Content-Type:
+      - application/x-protobuf
+      Date:
+      - Fri, 27 Dec 2024 22:14:53 GMT
+    status:
+      code: 200
+      message: OK
 - request:
     body: '{"messages": [{"role": "system", "content": "You are test role. test backstory\nYour
       personal goal is: test goal\nTo give my best complete final answer to the task
@@ -22,18 +105,20 @@ interactions:
       - '824'
       content-type:
       - application/json
+      cookie:
+      - _cfuvid=ePJSDFdHag2D8lj21_ijAMWjoA6xfnPNxN4uekvC728-1727226247743-0.0.1.1-604800000
       host:
       - api.openai.com
       user-agent:
       - OpenAI/Python 1.52.1
       x-stainless-arch:
-      - arm64
+      - x64
       x-stainless-async:
       - 'false'
       x-stainless-lang:
       - python
       x-stainless-os:
-      - MacOS
+      - Linux
       x-stainless-package-version:
       - 1.52.1
       x-stainless-raw-response:
@@ -47,8 +132,8 @@ interactions:
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-AaqIIsTxhvf75xvuu7gQScIlRSKbW\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1733344190,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+    content: "{\n  \"id\": \"chatcmpl-AjCtZLLrWi8ZASpP9bz6HaCV7xBIn\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1735337693,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
       \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
       Answer: Hi\",\n        \"refusal\": null\n      },\n      \"logprobs\": null,\n
@@ -57,12 +142,12 @@ interactions:
       {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
       {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
       0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_0705bf87c0\"\n}\n"
+      \"fp_0aa8d3e20b\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8ece8cfc3b1f4532-ATL
+      - 8f8caa83deca756b-SEA
       Connection:
       - keep-alive
       Content-Encoding:
@@ -70,14 +155,14 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 04 Dec 2024 20:29:50 GMT
+      - Fri, 27 Dec 2024 22:14:53 GMT
       Server:
       - cloudflare
       Set-Cookie:
-      - __cf_bm=QJZZjZ6eqnVamqUkw.Bx0mj7oBi3a_vGEH1VODcUxlg-1733344190-1.0.1.1-xyN0ekA9xIrSwEhRBmTiWJ3Pt72UYLU5owKfkz5yihVmMTfsr_Qz.ssGPJ5cuft066v1xVjb4zOSTdFmesMSKg;
-        path=/; expires=Wed, 04-Dec-24 20:59:50 GMT; domain=.api.openai.com; HttpOnly;
+      - __cf_bm=wJkq_yLkzE3OdxE0aMJz.G0kce969.9JxRmZ0ratl4c-1735337693-1.0.1.1-OKpUoRrSPFGvWv5Hp5ET1PNZ7iZNHPKEAuakpcQUxxPSeisUIIR3qIOZ31MGmYugqB5.wkvidgbxOAagqJvmnw;
+        path=/; expires=Fri, 27-Dec-24 22:44:53 GMT; domain=.api.openai.com; HttpOnly;
         Secure; SameSite=None
-      - _cfuvid=eCIkP8GVPvpkg19eOhCquWFHm.RTQBQy4yHLGGEAH5c-1733344190334-0.0.1.1-604800000;
+      - _cfuvid=A_ASCLNAVfQoyucWOAIhecWtEpNotYoZr0bAFihgNxs-1735337693273-0.0.1.1-604800000;
         path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
       Transfer-Encoding:
       - chunked
@@ -90,7 +175,7 @@ interactions:
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '313'
+      - '404'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -108,7 +193,7 @@ interactions:
       x-ratelimit-reset-tokens:
       - 0s
       x-request-id:
-      - req_9fd9a8ee688045dcf7ac5f6fdf689372
+      - req_6ac84634bff9193743c4b0911c09b4a6
     http_version: HTTP/1.1
     status_code: 200
 - request:
@@ -131,20 +216,20 @@ interactions:
       content-type:
       - application/json
       cookie:
-      - __cf_bm=QJZZjZ6eqnVamqUkw.Bx0mj7oBi3a_vGEH1VODcUxlg-1733344190-1.0.1.1-xyN0ekA9xIrSwEhRBmTiWJ3Pt72UYLU5owKfkz5yihVmMTfsr_Qz.ssGPJ5cuft066v1xVjb4zOSTdFmesMSKg;
-        _cfuvid=eCIkP8GVPvpkg19eOhCquWFHm.RTQBQy4yHLGGEAH5c-1733344190334-0.0.1.1-604800000
+      - _cfuvid=A_ASCLNAVfQoyucWOAIhecWtEpNotYoZr0bAFihgNxs-1735337693273-0.0.1.1-604800000;
+        __cf_bm=wJkq_yLkzE3OdxE0aMJz.G0kce969.9JxRmZ0ratl4c-1735337693-1.0.1.1-OKpUoRrSPFGvWv5Hp5ET1PNZ7iZNHPKEAuakpcQUxxPSeisUIIR3qIOZ31MGmYugqB5.wkvidgbxOAagqJvmnw
       host:
       - api.openai.com
       user-agent:
       - OpenAI/Python 1.52.1
       x-stainless-arch:
-      - arm64
+      - x64
       x-stainless-async:
       - 'false'
       x-stainless-lang:
       - python
       x-stainless-os:
-      - MacOS
+      - Linux
       x-stainless-package-version:
       - 1.52.1
       x-stainless-raw-response:
@@ -158,8 +243,8 @@ interactions:
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-AaqIIaQlLyoyPmk909PvAIfA2TmJL\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1733344190,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+    content: "{\n  \"id\": \"chatcmpl-AjCtZNlWdrrPZhq0MJDqd16sMuQEJ\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1735337693,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
       \"assistant\",\n        \"content\": \"True\",\n        \"refusal\": null\n
       \     },\n      \"logprobs\": null,\n      \"finish_reason\": \"stop\"\n    }\n
@@ -168,12 +253,12 @@ interactions:
       0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
       \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
       0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_0705bf87c0\"\n}\n"
+      \"fp_0aa8d3e20b\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8ece8d060b5e4532-ATL
+      - 8f8caa87094f756b-SEA
       Connection:
       - keep-alive
       Content-Encoding:
@@ -181,7 +266,7 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 04 Dec 2024 20:29:50 GMT
+      - Fri, 27 Dec 2024 22:14:53 GMT
       Server:
       - cloudflare
       Transfer-Encoding:
@@ -195,7 +280,7 @@ interactions:
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '375'
+      - '156'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -213,7 +298,7 @@ interactions:
       x-ratelimit-reset-tokens:
       - 0s
       x-request-id:
-      - req_be7cb475e0859a82c37ee3f2871ea5ea
+      - req_ec74bef2a9ef7b2144c03fd7f7bbeab0
     http_version: HTTP/1.1
     status_code: 200
 - request:
@@ -242,20 +327,20 @@ interactions:
       content-type:
       - application/json
       cookie:
-      - __cf_bm=QJZZjZ6eqnVamqUkw.Bx0mj7oBi3a_vGEH1VODcUxlg-1733344190-1.0.1.1-xyN0ekA9xIrSwEhRBmTiWJ3Pt72UYLU5owKfkz5yihVmMTfsr_Qz.ssGPJ5cuft066v1xVjb4zOSTdFmesMSKg;
-        _cfuvid=eCIkP8GVPvpkg19eOhCquWFHm.RTQBQy4yHLGGEAH5c-1733344190334-0.0.1.1-604800000
+      - _cfuvid=A_ASCLNAVfQoyucWOAIhecWtEpNotYoZr0bAFihgNxs-1735337693273-0.0.1.1-604800000;
+        __cf_bm=wJkq_yLkzE3OdxE0aMJz.G0kce969.9JxRmZ0ratl4c-1735337693-1.0.1.1-OKpUoRrSPFGvWv5Hp5ET1PNZ7iZNHPKEAuakpcQUxxPSeisUIIR3qIOZ31MGmYugqB5.wkvidgbxOAagqJvmnw
       host:
       - api.openai.com
       user-agent:
       - OpenAI/Python 1.52.1
       x-stainless-arch:
-      - arm64
+      - x64
       x-stainless-async:
       - 'false'
       x-stainless-lang:
       - python
       x-stainless-os:
-      - MacOS
+      - Linux
       x-stainless-package-version:
       - 1.52.1
       x-stainless-raw-response:
@@ -269,22 +354,23 @@ interactions:
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-AaqIJAAxpVfUOdrsgYKHwfRlHv4RS\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1733344191,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+    content: "{\n  \"id\": \"chatcmpl-AjCtZGv4f3h7GDdhyOy9G0sB1lRgC\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1735337693,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"Thought: I now can give a great answer
-      \ \\nFinal Answer: Hello\",\n        \"refusal\": null\n      },\n      \"logprobs\":
-      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-      188,\n    \"completion_tokens\": 14,\n    \"total_tokens\": 202,\n    \"prompt_tokens_details\":
-      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
-      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      \"assistant\",\n        \"content\": \"Thought: I understand the feedback and
+      will adjust my response accordingly.  \\nFinal Answer: Hello\",\n        \"refusal\":
+      null\n      },\n      \"logprobs\": null,\n      \"finish_reason\": \"stop\"\n
+      \   }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 188,\n    \"completion_tokens\":
+      18,\n    \"total_tokens\": 206,\n    \"prompt_tokens_details\": {\n      \"cached_tokens\":
+      0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
+      \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
       0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_0705bf87c0\"\n}\n"
+      \"fp_0aa8d3e20b\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8ece8d090fc34532-ATL
+      - 8f8caa88cac4756b-SEA
       Connection:
       - keep-alive
       Content-Encoding:
@@ -292,7 +378,7 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 04 Dec 2024 20:29:51 GMT
+      - Fri, 27 Dec 2024 22:14:54 GMT
       Server:
       - cloudflare
       Transfer-Encoding:
@@ -306,7 +392,7 @@ interactions:
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '484'
+      - '358'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -324,7 +410,7 @@ interactions:
       x-ratelimit-reset-tokens:
       - 0s
       x-request-id:
-      - req_5bf4a565ad6c2567a1ed204ecac89134
+      - req_ae1ab6b206d28ded6fee3c83ed0c2ab7
     http_version: HTTP/1.1
     status_code: 200
 - request:
@@ -346,20 +432,20 @@ interactions:
       content-type:
       - application/json
       cookie:
-      - __cf_bm=QJZZjZ6eqnVamqUkw.Bx0mj7oBi3a_vGEH1VODcUxlg-1733344190-1.0.1.1-xyN0ekA9xIrSwEhRBmTiWJ3Pt72UYLU5owKfkz5yihVmMTfsr_Qz.ssGPJ5cuft066v1xVjb4zOSTdFmesMSKg;
-        _cfuvid=eCIkP8GVPvpkg19eOhCquWFHm.RTQBQy4yHLGGEAH5c-1733344190334-0.0.1.1-604800000
+      - _cfuvid=A_ASCLNAVfQoyucWOAIhecWtEpNotYoZr0bAFihgNxs-1735337693273-0.0.1.1-604800000;
+        __cf_bm=wJkq_yLkzE3OdxE0aMJz.G0kce969.9JxRmZ0ratl4c-1735337693-1.0.1.1-OKpUoRrSPFGvWv5Hp5ET1PNZ7iZNHPKEAuakpcQUxxPSeisUIIR3qIOZ31MGmYugqB5.wkvidgbxOAagqJvmnw
       host:
       - api.openai.com
       user-agent:
       - OpenAI/Python 1.52.1
       x-stainless-arch:
-      - arm64
+      - x64
       x-stainless-async:
       - 'false'
       x-stainless-lang:
       - python
       x-stainless-os:
-      - MacOS
+      - Linux
       x-stainless-package-version:
       - 1.52.1
       x-stainless-raw-response:
@@ -373,8 +459,8 @@ interactions:
     method: POST
     uri: https://api.openai.com/v1/chat/completions
   response:
-    content: "{\n  \"id\": \"chatcmpl-AaqIJqyG8vl9mxj2qDPZgaxyNLLIq\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1733344191,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+    content: "{\n  \"id\": \"chatcmpl-AjCtaiHL4TY8Dssk0j2miqmjrzquy\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1735337694,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
       \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
       \"assistant\",\n        \"content\": \"False\",\n        \"refusal\": null\n
       \     },\n      \"logprobs\": null,\n      \"finish_reason\": \"stop\"\n    }\n
@@ -383,12 +469,12 @@ interactions:
       0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
       \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
       0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_0705bf87c0\"\n}\n"
+      \"fp_0aa8d3e20b\"\n}\n"
     headers:
       CF-Cache-Status:
       - DYNAMIC
       CF-RAY:
-      - 8ece8d0cfdeb4532-ATL
+      - 8f8caa8bdd26756b-SEA
       Connection:
       - keep-alive
       Content-Encoding:
@@ -396,7 +482,7 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Wed, 04 Dec 2024 20:29:51 GMT
+      - Fri, 27 Dec 2024 22:14:54 GMT
       Server:
       - cloudflare
       Transfer-Encoding:
@@ -410,7 +496,7 @@ interactions:
       openai-organization:
       - crewai-iuxna1
       openai-processing-ms:
-      - '341'
+      - '184'
       openai-version:
       - '2020-10-01'
       strict-transport-security:
@@ -428,7 +514,7 @@ interactions:
       x-ratelimit-reset-tokens:
       - 0s
       x-request-id:
-      - req_5554bade8ceda00cf364b76a51b708ff
+      - req_652891f79c1104a7a8436275d78a69f1
     http_version: HTTP/1.1
     status_code: 200
 version: 1

tests/cassettes/test_agent_tool_role_matching[ "Futel Official Infopoint" -True].yaml

@@ -1,117 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Futel Official Infopoint.
-      Futel Football Club info\nYour personal goal is: Answer questions about Futel\nTo
-      give my best complete final answer to the task respond using the exact following
-      format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
-      answer must be the great and the most complete as possible, it must be outcome
-      described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
-      "content": "\nCurrent Task: Test task\n\nThis is the expect criteria for your
-      final answer: Your best answer to your coworker asking you this, accounting
-      for the context shared.\nyou MUST return the actual complete content as the
-      final answer, not a summary.\n\nBegin! This is VERY important to you, use the
-      tools available and give your best Final Answer, your job depends on it!\n\nThought:"}],
-      "model": "gpt-4o", "stop": ["\nObservation:"], "stream": false}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '939'
-      content-type:
-      - application/json
-      cookie:
-      - __cf_bm=cwWdOaPJjFMNJaLtJfa8Kjqavswg5bzVRFzBX4gneGw-1736458417-1.0.1.1-bvf2HshgcMtgn7GdxqwySFDAIacGccDFfEXniBFTTDmbGMCiIIwf6t2DiwWnBldmUHixwc5kDO9gYs08g.feBA;
-        _cfuvid=WMw7PSqkYqQOieguBRs0uNkwNU92A.ZKbgDbCAcV3EQ-1736458417825-0.0.1.1-604800000
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.52.1
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.52.1
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-retry-count:
-      - '0'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.12.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AnuRlxiTxduAVoXHHY58Fvfbll5IS\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1736458417,\n  \"model\": \"gpt-4o-2024-08-06\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
-      Answer: This is a test task, and the context or question from the coworker is
-      not specified. Therefore, my best effort would be to affirm my readiness to
-      answer accurately and in detail any question about Futel Football Club based
-      on the context described. If provided with specific information or questions,
-      I will ensure to respond comprehensively as required by my job directives.\",\n
-      \       \"refusal\": null\n      },\n      \"logprobs\": null,\n      \"finish_reason\":
-      \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 177,\n    \"completion_tokens\":
-      82,\n    \"total_tokens\": 259,\n    \"prompt_tokens_details\": {\n      \"cached_tokens\":
-      0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
-      \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
-      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_703d4ff298\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8ff78bf7bd6cc002-ATL
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Thu, 09 Jan 2025 21:33:40 GMT
-      Server:
-      - cloudflare
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      alt-svc:
-      - h3=":443"; ma=86400
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '2263'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '10000'
-      x-ratelimit-limit-tokens:
-      - '30000000'
-      x-ratelimit-remaining-requests:
-      - '9999'
-      x-ratelimit-remaining-tokens:
-      - '29999786'
-      x-ratelimit-reset-requests:
-      - 6ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_7c1a31da73cd103e9f410f908e59187f
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_agent_tool_role_matching[ FUTEL\nOFFICIAL INFOPOINT -True].yaml

@@ -1,119 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Futel Official Infopoint.
-      Futel Football Club info\nYour personal goal is: Answer questions about Futel\nTo
-      give my best complete final answer to the task respond using the exact following
-      format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
-      answer must be the great and the most complete as possible, it must be outcome
-      described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
-      "content": "\nCurrent Task: Test task\n\nThis is the expect criteria for your
-      final answer: Your best answer to your coworker asking you this, accounting
-      for the context shared.\nyou MUST return the actual complete content as the
-      final answer, not a summary.\n\nBegin! This is VERY important to you, use the
-      tools available and give your best Final Answer, your job depends on it!\n\nThought:"}],
-      "model": "gpt-4o", "stop": ["\nObservation:"], "stream": false}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '939'
-      content-type:
-      - application/json
-      cookie:
-      - __cf_bm=cwWdOaPJjFMNJaLtJfa8Kjqavswg5bzVRFzBX4gneGw-1736458417-1.0.1.1-bvf2HshgcMtgn7GdxqwySFDAIacGccDFfEXniBFTTDmbGMCiIIwf6t2DiwWnBldmUHixwc5kDO9gYs08g.feBA;
-        _cfuvid=WMw7PSqkYqQOieguBRs0uNkwNU92A.ZKbgDbCAcV3EQ-1736458417825-0.0.1.1-604800000
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.52.1
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.52.1
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-retry-count:
-      - '0'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.12.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AnuRrFJZGKw8cIEshvuW1PKwFZFKs\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1736458423,\n  \"model\": \"gpt-4o-2024-08-06\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
-      Answer: Although you mentioned this being a \\\"Test task\\\" and haven't provided
-      a specific question regarding Futel Football Club, your request appears to involve
-      ensuring accuracy and detail in responses. For a proper answer about Futel,
-      I'd be ready to provide details about the club's history, management, players,
-      match schedules, and recent performance statistics. Remember to ask specific
-      questions to receive a targeted response. If this were a real context where
-      information was shared, I would respond precisely to what's been asked regarding
-      Futel Football Club.\",\n        \"refusal\": null\n      },\n      \"logprobs\":
-      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-      177,\n    \"completion_tokens\": 113,\n    \"total_tokens\": 290,\n    \"prompt_tokens_details\":
-      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
-      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
-      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_703d4ff298\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8ff78c1d0ecdc002-ATL
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Thu, 09 Jan 2025 21:33:47 GMT
-      Server:
-      - cloudflare
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      alt-svc:
-      - h3=":443"; ma=86400
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '3097'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '10000'
-      x-ratelimit-limit-tokens:
-      - '30000000'
-      x-ratelimit-remaining-requests:
-      - '9999'
-      x-ratelimit-remaining-tokens:
-      - '29999786'
-      x-ratelimit-reset-requests:
-      - 6ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_179e1d56e2b17303e40480baffbc7b08
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_agent_tool_role_matching["Futel Official Infopoint"-True].yaml

@@ -1,114 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Futel Official Infopoint.
-      Futel Football Club info\nYour personal goal is: Answer questions about Futel\nTo
-      give my best complete final answer to the task respond using the exact following
-      format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
-      answer must be the great and the most complete as possible, it must be outcome
-      described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
-      "content": "\nCurrent Task: Test task\n\nThis is the expect criteria for your
-      final answer: Your best answer to your coworker asking you this, accounting
-      for the context shared.\nyou MUST return the actual complete content as the
-      final answer, not a summary.\n\nBegin! This is VERY important to you, use the
-      tools available and give your best Final Answer, your job depends on it!\n\nThought:"}],
-      "model": "gpt-4o", "stop": ["\nObservation:"], "stream": false}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '939'
-      content-type:
-      - application/json
-      cookie:
-      - __cf_bm=cwWdOaPJjFMNJaLtJfa8Kjqavswg5bzVRFzBX4gneGw-1736458417-1.0.1.1-bvf2HshgcMtgn7GdxqwySFDAIacGccDFfEXniBFTTDmbGMCiIIwf6t2DiwWnBldmUHixwc5kDO9gYs08g.feBA;
-        _cfuvid=WMw7PSqkYqQOieguBRs0uNkwNU92A.ZKbgDbCAcV3EQ-1736458417825-0.0.1.1-604800000
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.52.1
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.52.1
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-retry-count:
-      - '0'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.12.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AnuRqgg7eiHnDi2DOqdk99fiqOboz\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1736458422,\n  \"model\": \"gpt-4o-2024-08-06\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
-      Answer: Your best answer to your coworker asking you this, accounting for the
-      context shared. You MUST return the actual complete content as the final answer,
-      not a summary.\",\n        \"refusal\": null\n      },\n      \"logprobs\":
-      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-      177,\n    \"completion_tokens\": 44,\n    \"total_tokens\": 221,\n    \"prompt_tokens_details\":
-      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
-      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
-      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_703d4ff298\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8ff78c164ad2c002-ATL
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Thu, 09 Jan 2025 21:33:43 GMT
-      Server:
-      - cloudflare
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      alt-svc:
-      - h3=":443"; ma=86400
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '899'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '10000'
-      x-ratelimit-limit-tokens:
-      - '30000000'
-      x-ratelimit-remaining-requests:
-      - '9999'
-      x-ratelimit-remaining-tokens:
-      - '29999786'
-      x-ratelimit-reset-requests:
-      - 6ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_9f5226208edb90a27987aaf7e0ca03d3
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_agent_tool_role_matching[Futel Official Infopoint-True].yaml

@@ -1,119 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Futel Official Infopoint.
-      Futel Football Club info\nYour personal goal is: Answer questions about Futel\nTo
-      give my best complete final answer to the task respond using the exact following
-      format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
-      answer must be the great and the most complete as possible, it must be outcome
-      described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
-      "content": "\nCurrent Task: Test task\n\nThis is the expect criteria for your
-      final answer: Your best answer to your coworker asking you this, accounting
-      for the context shared.\nyou MUST return the actual complete content as the
-      final answer, not a summary.\n\nBegin! This is VERY important to you, use the
-      tools available and give your best Final Answer, your job depends on it!\n\nThought:"}],
-      "model": "gpt-4o", "stop": ["\nObservation:"], "stream": false}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '939'
-      content-type:
-      - application/json
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.52.1
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.52.1
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-retry-count:
-      - '0'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.12.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AnuRjmwH5mrykLxQhFwTqqTiDtuTf\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1736458415,\n  \"model\": \"gpt-4o-2024-08-06\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
-      Answer: As this is a test task, please note that Futel Football Club is fictional
-      and any specific details about it would not be available. However, if you have
-      specific questions or need information about a particular aspect of Futel or
-      any general football club inquiry, feel free to ask, and I'll do my best to
-      assist you with your query!\",\n        \"refusal\": null\n      },\n      \"logprobs\":
-      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-      177,\n    \"completion_tokens\": 79,\n    \"total_tokens\": 256,\n    \"prompt_tokens_details\":
-      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
-      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
-      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_703d4ff298\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8ff78be5eebfc002-ATL
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Thu, 09 Jan 2025 21:33:37 GMT
-      Server:
-      - cloudflare
-      Set-Cookie:
-      - __cf_bm=cwWdOaPJjFMNJaLtJfa8Kjqavswg5bzVRFzBX4gneGw-1736458417-1.0.1.1-bvf2HshgcMtgn7GdxqwySFDAIacGccDFfEXniBFTTDmbGMCiIIwf6t2DiwWnBldmUHixwc5kDO9gYs08g.feBA;
-        path=/; expires=Thu, 09-Jan-25 22:03:37 GMT; domain=.api.openai.com; HttpOnly;
-        Secure; SameSite=None
-      - _cfuvid=WMw7PSqkYqQOieguBRs0uNkwNU92A.ZKbgDbCAcV3EQ-1736458417825-0.0.1.1-604800000;
-        path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      alt-svc:
-      - h3=":443"; ma=86400
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '2730'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '10000'
-      x-ratelimit-limit-tokens:
-      - '30000000'
-      x-ratelimit-remaining-requests:
-      - '9999'
-      x-ratelimit-remaining-tokens:
-      - '29999786'
-      x-ratelimit-reset-requests:
-      - 6ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_014478ba748f860d10ac250ca0ba824a
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_agent_tool_role_matching[Futel Official Infopoint\n-True].yaml

@@ -1,119 +0,0 @@
-interactions:
-- request:
-    body: '{"messages": [{"role": "system", "content": "You are Futel Official Infopoint.
-      Futel Football Club info\nYour personal goal is: Answer questions about Futel\nTo
-      give my best complete final answer to the task respond using the exact following
-      format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
-      answer must be the great and the most complete as possible, it must be outcome
-      described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
-      "content": "\nCurrent Task: Test task\n\nThis is the expect criteria for your
-      final answer: Your best answer to your coworker asking you this, accounting
-      for the context shared.\nyou MUST return the actual complete content as the
-      final answer, not a summary.\n\nBegin! This is VERY important to you, use the
-      tools available and give your best Final Answer, your job depends on it!\n\nThought:"}],
-      "model": "gpt-4o", "stop": ["\nObservation:"], "stream": false}'
-    headers:
-      accept:
-      - application/json
-      accept-encoding:
-      - gzip, deflate
-      connection:
-      - keep-alive
-      content-length:
-      - '939'
-      content-type:
-      - application/json
-      cookie:
-      - __cf_bm=cwWdOaPJjFMNJaLtJfa8Kjqavswg5bzVRFzBX4gneGw-1736458417-1.0.1.1-bvf2HshgcMtgn7GdxqwySFDAIacGccDFfEXniBFTTDmbGMCiIIwf6t2DiwWnBldmUHixwc5kDO9gYs08g.feBA;
-        _cfuvid=WMw7PSqkYqQOieguBRs0uNkwNU92A.ZKbgDbCAcV3EQ-1736458417825-0.0.1.1-604800000
-      host:
-      - api.openai.com
-      user-agent:
-      - OpenAI/Python 1.52.1
-      x-stainless-arch:
-      - arm64
-      x-stainless-async:
-      - 'false'
-      x-stainless-lang:
-      - python
-      x-stainless-os:
-      - MacOS
-      x-stainless-package-version:
-      - 1.52.1
-      x-stainless-raw-response:
-      - 'true'
-      x-stainless-retry-count:
-      - '0'
-      x-stainless-runtime:
-      - CPython
-      x-stainless-runtime-version:
-      - 3.12.7
-    method: POST
-    uri: https://api.openai.com/v1/chat/completions
-  response:
-    content: "{\n  \"id\": \"chatcmpl-AnuRofLgmzWcDya5LILqYwIJYgFoq\",\n  \"object\":
-      \"chat.completion\",\n  \"created\": 1736458420,\n  \"model\": \"gpt-4o-2024-08-06\",\n
-      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-      \"assistant\",\n        \"content\": \"I now can give a great answer  \\nFinal
-      Answer: As an official Futel Football Club infopoint, my responsibility is to
-      provide detailed and accurate information about the club. This includes answering
-      questions regarding team statistics, player performances, upcoming fixtures,
-      ticketing and fan zone details, club history, and community initiatives. Our
-      focus is to ensure that fans and stakeholders have access to the latest and
-      most precise information about the club's on and off-pitch activities. If there's
-      anything specific you need to know, just let me know, and I'll be more than
-      happy to assist!\",\n        \"refusal\": null\n      },\n      \"logprobs\":
-      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-      177,\n    \"completion_tokens\": 115,\n    \"total_tokens\": 292,\n    \"prompt_tokens_details\":
-      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
-      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
-      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"system_fingerprint\":
-      \"fp_703d4ff298\"\n}\n"
-    headers:
-      CF-Cache-Status:
-      - DYNAMIC
-      CF-RAY:
-      - 8ff78c066f37c002-ATL
-      Connection:
-      - keep-alive
-      Content-Encoding:
-      - gzip
-      Content-Type:
-      - application/json
-      Date:
-      - Thu, 09 Jan 2025 21:33:42 GMT
-      Server:
-      - cloudflare
-      Transfer-Encoding:
-      - chunked
-      X-Content-Type-Options:
-      - nosniff
-      access-control-expose-headers:
-      - X-Request-ID
-      alt-svc:
-      - h3=":443"; ma=86400
-      openai-organization:
-      - crewai-iuxna1
-      openai-processing-ms:
-      - '2459'
-      openai-version:
-      - '2020-10-01'
-      strict-transport-security:
-      - max-age=31536000; includeSubDomains; preload
-      x-ratelimit-limit-requests:
-      - '10000'
-      x-ratelimit-limit-tokens:
-      - '30000000'
-      x-ratelimit-remaining-requests:
-      - '9999'
-      x-ratelimit-remaining-tokens:
-      - '29999786'
-      x-ratelimit-reset-requests:
-      - 6ms
-      x-ratelimit-reset-tokens:
-      - 0s
-      x-request-id:
-      - req_a146dd27f040f39a576750970cca0f52
-    http_version: HTTP/1.1
-    status_code: 200
-version: 1

tests/cassettes/test_deepseek_r1_with_open_router.yaml

@@ -0,0 +1,100 @@
+interactions:
+- request:
+    body: '{"model": "deepseek/deepseek-r1", "messages": [{"role": "user", "content":
+      "What is the capital of France?"}], "stop": [], "stream": false}'
+    headers:
+      accept:
+      - '*/*'
+      accept-encoding:
+      - gzip, deflate
+      connection:
+      - keep-alive
+      content-length:
+      - '139'
+      host:
+      - openrouter.ai
+      http-referer:
+      - https://litellm.ai
+      user-agent:
+      - litellm/1.60.2
+      x-title:
+      - liteLLM
+    method: POST
+    uri: https://openrouter.ai/api/v1/chat/completions
+  response:
+    content: "\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n\n
+      \        \n\n         \n\n         \n\n         \n\n         \n\n         \n{\"id\":\"gen-1738684300-YnD5WOSczQWsW0vQG78a\",\"provider\":\"Nebius\",\"model\":\"deepseek/deepseek-r1\",\"object\":\"chat.completion\",\"created\":1738684300,\"choices\":[{\"logprobs\":null,\"index\":0,\"message\":{\"role\":\"assistant\",\"content\":\"The
+      capital of France is **Paris**. Known for its iconic landmarks such as the Eiffel
+      Tower, Notre-Dame Cathedral, and the Louvre Museum, Paris has served as the
+      political and cultural center of France for centuries. \U0001F1EB\U0001F1F7\",\"refusal\":null}}],\"usage\":{\"prompt_tokens\":10,\"completion_tokens\":261,\"total_tokens\":271}}"
+    headers:
+      Access-Control-Allow-Origin:
+      - '*'
+      CF-RAY:
+      - 90cbd2ceaf3ead5e-ATL
+      Connection:
+      - keep-alive
+      Content-Encoding:
+      - gzip
+      Content-Type:
+      - application/json
+      Date:
+      - Tue, 04 Feb 2025 15:51:40 GMT
+      Server:
+      - cloudflare
+      Transfer-Encoding:
+      - chunked
+      Vary:
+      - Accept-Encoding
+      x-clerk-auth-message:
+      - Invalid JWT form. A JWT consists of three parts separated by dots. (reason=token-invalid,
+        token-carrier=header)
+      x-clerk-auth-reason:
+      - token-invalid
+      x-clerk-auth-status:
+      - signed-out
+    http_version: HTTP/1.1
+    status_code: 200
+version: 1

tests/cassettes/test_llm_call_with_message_list.yaml

@@ -0,0 +1,102 @@
+interactions:
+- request:
+    body: '{"messages": [{"role": "user", "content": "What is the capital of France?"}],
+      "model": "gpt-4o-mini"}'
+    headers:
+      accept:
+      - application/json
+      accept-encoding:
+      - gzip, deflate
+      connection:
+      - keep-alive
+      content-length:
+      - '101'
+      content-type:
+      - application/json
+      cookie:
+      - _cfuvid=8NrWEBP3dDmc8p2.csR.EdsSwS8zFvzWI1kPICaK_fM-1737568015338-0.0.1.1-604800000;
+        __cf_bm=pKr3NwXmTZN9rMSlKvEX40VPKbrxF93QwDNHunL2v8Y-1737568015-1.0.1.1-nR0EA7hYIwWpIBYUI53d9xQrUnl5iML6lgz4AGJW4ZGPBDxFma3PZ2cBhlr_hE7wKa5fV3r32eMu_rNWMXD.eA
+      host:
+      - api.openai.com
+      user-agent:
+      - OpenAI/Python 1.59.6
+      x-stainless-arch:
+      - arm64
+      x-stainless-async:
+      - 'false'
+      x-stainless-lang:
+      - python
+      x-stainless-os:
+      - MacOS
+      x-stainless-package-version:
+      - 1.59.6
+      x-stainless-raw-response:
+      - 'true'
+      x-stainless-retry-count:
+      - '0'
+      x-stainless-runtime:
+      - CPython
+      x-stainless-runtime-version:
+      - 3.12.7
+    method: POST
+    uri: https://api.openai.com/v1/chat/completions
+  response:
+    content: "{\n  \"id\": \"chatcmpl-AsZ6WjNfEOrHwwEEdSZZCRBiTpBMS\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1737568016,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
+      \"assistant\",\n        \"content\": \"The capital of France is Paris.\",\n
+      \       \"refusal\": null\n      },\n      \"logprobs\": null,\n      \"finish_reason\":
+      \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 14,\n    \"completion_tokens\":
+      8,\n    \"total_tokens\": 22,\n    \"prompt_tokens_details\": {\n      \"cached_tokens\":
+      0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
+      \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"service_tier\":
+      \"default\",\n  \"system_fingerprint\": \"fp_72ed7ab54c\"\n}\n"
+    headers:
+      CF-Cache-Status:
+      - DYNAMIC
+      CF-RAY:
+      - 90615dc63b805cb1-RDU
+      Connection:
+      - keep-alive
+      Content-Encoding:
+      - gzip
+      Content-Type:
+      - application/json
+      Date:
+      - Wed, 22 Jan 2025 17:46:56 GMT
+      Server:
+      - cloudflare
+      Transfer-Encoding:
+      - chunked
+      X-Content-Type-Options:
+      - nosniff
+      access-control-expose-headers:
+      - X-Request-ID
+      alt-svc:
+      - h3=":443"; ma=86400
+      openai-organization:
+      - crewai-iuxna1
+      openai-processing-ms:
+      - '355'
+      openai-version:
+      - '2020-10-01'
+      strict-transport-security:
+      - max-age=31536000; includeSubDomains; preload
+      x-ratelimit-limit-requests:
+      - '30000'
+      x-ratelimit-limit-tokens:
+      - '150000000'
+      x-ratelimit-remaining-requests:
+      - '29999'
+      x-ratelimit-remaining-tokens:
+      - '149999974'
+      x-ratelimit-reset-requests:
+      - 2ms
+      x-ratelimit-reset-tokens:
+      - 0s
+      x-request-id:
+      - req_cdbed69c9c63658eb552b07f1220df19
+    http_version: HTTP/1.1
+    status_code: 200
+version: 1

tests/cassettes/test_llm_call_with_string_input.yaml

@@ -0,0 +1,108 @@
+interactions:
+- request:
+    body: '{"messages": [{"role": "user", "content": "Return the name of a random
+      city in the world."}], "model": "gpt-4o-mini"}'
+    headers:
+      accept:
+      - application/json
+      accept-encoding:
+      - gzip, deflate
+      connection:
+      - keep-alive
+      content-length:
+      - '117'
+      content-type:
+      - application/json
+      cookie:
+      - _cfuvid=3UeEmz_rnmsoZxrVUv32u35gJOi766GDWNe5_RTjiPk-1736537376739-0.0.1.1-604800000
+      host:
+      - api.openai.com
+      user-agent:
+      - OpenAI/Python 1.59.6
+      x-stainless-arch:
+      - arm64
+      x-stainless-async:
+      - 'false'
+      x-stainless-lang:
+      - python
+      x-stainless-os:
+      - MacOS
+      x-stainless-package-version:
+      - 1.59.6
+      x-stainless-raw-response:
+      - 'true'
+      x-stainless-retry-count:
+      - '0'
+      x-stainless-runtime:
+      - CPython
+      x-stainless-runtime-version:
+      - 3.12.7
+    method: POST
+    uri: https://api.openai.com/v1/chat/completions
+  response:
+    content: "{\n  \"id\": \"chatcmpl-AsZ6UtbaNSMpNU9VJKxvn52t5eJTq\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1737568014,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
+      \"assistant\",\n        \"content\": \"How about \\\"Lisbon\\\"? It\u2019s the
+      capital city of Portugal, known for its rich history and vibrant culture.\",\n
+      \       \"refusal\": null\n      },\n      \"logprobs\": null,\n      \"finish_reason\":
+      \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\": 18,\n    \"completion_tokens\":
+      24,\n    \"total_tokens\": 42,\n    \"prompt_tokens_details\": {\n      \"cached_tokens\":
+      0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\": {\n
+      \     \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"service_tier\":
+      \"default\",\n  \"system_fingerprint\": \"fp_72ed7ab54c\"\n}\n"
+    headers:
+      CF-Cache-Status:
+      - DYNAMIC
+      CF-RAY:
+      - 90615dbcaefb5cb1-RDU
+      Connection:
+      - keep-alive
+      Content-Encoding:
+      - gzip
+      Content-Type:
+      - application/json
+      Date:
+      - Wed, 22 Jan 2025 17:46:55 GMT
+      Server:
+      - cloudflare
+      Set-Cookie:
+      - __cf_bm=pKr3NwXmTZN9rMSlKvEX40VPKbrxF93QwDNHunL2v8Y-1737568015-1.0.1.1-nR0EA7hYIwWpIBYUI53d9xQrUnl5iML6lgz4AGJW4ZGPBDxFma3PZ2cBhlr_hE7wKa5fV3r32eMu_rNWMXD.eA;
+        path=/; expires=Wed, 22-Jan-25 18:16:55 GMT; domain=.api.openai.com; HttpOnly;
+        Secure; SameSite=None
+      - _cfuvid=8NrWEBP3dDmc8p2.csR.EdsSwS8zFvzWI1kPICaK_fM-1737568015338-0.0.1.1-604800000;
+        path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
+      Transfer-Encoding:
+      - chunked
+      X-Content-Type-Options:
+      - nosniff
+      access-control-expose-headers:
+      - X-Request-ID
+      alt-svc:
+      - h3=":443"; ma=86400
+      openai-organization:
+      - crewai-iuxna1
+      openai-processing-ms:
+      - '449'
+      openai-version:
+      - '2020-10-01'
+      strict-transport-security:
+      - max-age=31536000; includeSubDomains; preload
+      x-ratelimit-limit-requests:
+      - '30000'
+      x-ratelimit-limit-tokens:
+      - '150000000'
+      x-ratelimit-remaining-requests:
+      - '29999'
+      x-ratelimit-remaining-tokens:
+      - '149999971'
+      x-ratelimit-reset-requests:
+      - 2ms
+      x-ratelimit-reset-tokens:
+      - 0s
+      x-request-id:
+      - req_898373758d2eae3cd84814050b2588e3
+    http_version: HTTP/1.1
+    status_code: 200
+version: 1

tests/cassettes/test_llm_call_with_string_input_and_callbacks.yaml

@@ -0,0 +1,102 @@
+interactions:
+- request:
+    body: '{"messages": [{"role": "user", "content": "Tell me a joke."}], "model":
+      "gpt-4o-mini"}'
+    headers:
+      accept:
+      - application/json
+      accept-encoding:
+      - gzip, deflate
+      connection:
+      - keep-alive
+      content-length:
+      - '86'
+      content-type:
+      - application/json
+      cookie:
+      - _cfuvid=8NrWEBP3dDmc8p2.csR.EdsSwS8zFvzWI1kPICaK_fM-1737568015338-0.0.1.1-604800000;
+        __cf_bm=pKr3NwXmTZN9rMSlKvEX40VPKbrxF93QwDNHunL2v8Y-1737568015-1.0.1.1-nR0EA7hYIwWpIBYUI53d9xQrUnl5iML6lgz4AGJW4ZGPBDxFma3PZ2cBhlr_hE7wKa5fV3r32eMu_rNWMXD.eA
+      host:
+      - api.openai.com
+      user-agent:
+      - OpenAI/Python 1.59.6
+      x-stainless-arch:
+      - arm64
+      x-stainless-async:
+      - 'false'
+      x-stainless-lang:
+      - python
+      x-stainless-os:
+      - MacOS
+      x-stainless-package-version:
+      - 1.59.6
+      x-stainless-raw-response:
+      - 'true'
+      x-stainless-retry-count:
+      - '0'
+      x-stainless-runtime:
+      - CPython
+      x-stainless-runtime-version:
+      - 3.12.7
+    method: POST
+    uri: https://api.openai.com/v1/chat/completions
+  response:
+    content: "{\n  \"id\": \"chatcmpl-AsZ6VyjuUcXYpChXmD8rUSy6nSGq8\",\n  \"object\":
+      \"chat.completion\",\n  \"created\": 1737568015,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
+      \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
+      \"assistant\",\n        \"content\": \"Why did the scarecrow win an award?  \\n\\nBecause
+      he was outstanding in his field!\",\n        \"refusal\": null\n      },\n      \"logprobs\":
+      null,\n      \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
+      12,\n    \"completion_tokens\": 19,\n    \"total_tokens\": 31,\n    \"prompt_tokens_details\":
+      {\n      \"cached_tokens\": 0,\n      \"audio_tokens\": 0\n    },\n    \"completion_tokens_details\":
+      {\n      \"reasoning_tokens\": 0,\n      \"audio_tokens\": 0,\n      \"accepted_prediction_tokens\":
+      0,\n      \"rejected_prediction_tokens\": 0\n    }\n  },\n  \"service_tier\":
+      \"default\",\n  \"system_fingerprint\": \"fp_72ed7ab54c\"\n}\n"
+    headers:
+      CF-Cache-Status:
+      - DYNAMIC
+      CF-RAY:
+      - 90615dc03b6c5cb1-RDU
+      Connection:
+      - keep-alive
+      Content-Encoding:
+      - gzip
+      Content-Type:
+      - application/json
+      Date:
+      - Wed, 22 Jan 2025 17:46:56 GMT
+      Server:
+      - cloudflare
+      Transfer-Encoding:
+      - chunked
+      X-Content-Type-Options:
+      - nosniff
+      access-control-expose-headers:
+      - X-Request-ID
+      alt-svc:
+      - h3=":443"; ma=86400
+      openai-organization:
+      - crewai-iuxna1
+      openai-processing-ms:
+      - '825'
+      openai-version:
+      - '2020-10-01'
+      strict-transport-security:
+      - max-age=31536000; includeSubDomains; preload
+      x-ratelimit-limit-requests:
+      - '30000'
+      x-ratelimit-limit-tokens:
+      - '150000000'
+      x-ratelimit-remaining-requests:
+      - '29999'
+      x-ratelimit-remaining-tokens:
+      - '149999979'
+      x-ratelimit-reset-requests:
+      - 2ms
+      x-ratelimit-reset-tokens:
+      - 0s
+      x-request-id:
+      - req_4c1485d44e7461396d4a7316a63ff353
+    http_version: HTTP/1.1
+    status_code: 200
+version: 1