Compare commits

...

20 Commits

Author SHA1 Message Date
Devin AI
8ab236a68e Fix lint issues: remove unused imports
- Remove unused 'Crew' import from both test files
- Remove unused 'pytest' import from test_langdb_documentation.py
- Keep only imports that are actually used in the code

Fixes lint check failure in PR #3241

Co-Authored-By: João <joao@crewai.com>
2025-07-30 10:06:59 +00:00
Devin AI
dce11df0b7 Add LangDB AI Gateway documentation to observability section
- Add comprehensive LangDB documentation following Portkey pattern
- Include installation, configuration, and integration examples
- Add LangDB card to observability overview page
- Include tests for documentation examples
- Addresses issue #3240: Feature request for LangDB observability docs

Features documented:
- Complete end-to-end tracing of agent interactions
- Real-time cost monitoring and optimization
- Performance analytics with detailed metrics
- Enterprise security and governance features
- Multi-environment setup configurations
- Advanced metadata and filtering capabilities

Co-Authored-By: João <joao@crewai.com>
2025-07-30 10:02:49 +00:00
Lorenze Jay
cb522cf500 Enhance Flow class to support custom flow names (#3234)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
- Added an optional `name` attribute to the Flow class for better identification.
- Updated event emissions to utilize the new `name` attribute, ensuring accurate flow naming in events.
- Added tests to verify the correct flow name is set and emitted during flow execution.
2025-07-29 15:41:30 -07:00
Vini Brasil
017acc74f5 Add timezone to event timestamps (#3231)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
Events were lacking timezone information, making them naive datetimes,
which can be ambiguous.
2025-07-28 17:09:06 -03:00
Greyson LaLonde
fab86d197a Refactor: Move RAG components to dedicated top-level module (#3222)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
* Move RAG components to top-level module

- Create src/crewai/rag directory structure
- Move embeddings configurator from utilities to rag module
- Update imports across codebase and documentation
- Remove deprecated embedding files

* Remove empty knowledge/embedder directory
2025-07-25 10:55:31 -04:00
Vidit Ostwal
864e9bfb76 Changed the default value in Mem0 config (#3216)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
* Changed the default value in Mem0 config

* Added regression test for this

* Fixed Linting issues
2025-07-24 13:20:18 -04:00
Lucas Gomide
d3b45d197c fix: remove crewai signup references, replaced by crewai login (#3213) 2025-07-24 07:47:35 -04:00
Manuka Yasas
579153b070 docs: fix incorrect model naming in Google Vertex AI documentation (#3189)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
- Change model format from "gemini/gemini-1.5-pro-latest" to "gemini-1.5-pro-latest"
  in Vertex AI section examples
- Update both English and Portuguese documentation files
- Fixes incorrect provider prefix usage for Vertex AI models
- Ensures consistency with Vertex AI provider requirements

Files changed:
- docs/en/concepts/llms.mdx (line 272)
- docs/pt-BR/concepts/llms.mdx (line 270)

Co-authored-by: Tony Kipkemboi <iamtonykipkemboi@gmail.com>
2025-07-23 16:58:57 -04:00
Lorenze Jay
b1fdcdfa6e chore: update dependencies and version in project files (#3212)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
- Updated `crewai-tools` dependency from `0.55.0` to `0.58.0` in `pyproject.toml` and `uv.lock`.
- Added new packages `anthropic`, `browserbase`, `playwright`, `pyee`, and `stagehand` with their respective versions in `uv.lock`.
- Bumped the version of the CrewAI library from `0.148.0` to `0.150.0` in `__init__.py`.
- Updated dependency versions in CLI templates for crew, flow, and tool projects to reflect the new CrewAI version.
2025-07-23 11:03:50 -07:00
Mike Plachta
18d76a270c docs: add SerperScrapeWebsiteTool documentation and reorganize SerperDevTool setup instructions (#3211) 2025-07-23 12:12:59 -04:00
Vidit Ostwal
30541239ad Changed Mem0 Storage v1.1 -> v2 (#2893)
* Changed v1.1 -> v2

* Fixed Test Cases:

* Fixed linting issues

* Changed docs

* Refractored the storage

* Fixed test cases

* Fixing run-time checks

* Fixed Test Case

* Updated docs and added test case for custom categories

* Add the TODO back

* Minor Changes

* Added output_format in search

* Minor changes

* Added output_format and version in both search and save

* Small change

* Minor bugs

* Fixed test cases

* Changed docs

---------

Co-authored-by: Lucas Gomide <lucaslg200@gmail.com>
2025-07-23 08:30:52 -04:00
Tony Kipkemboi
9a65573955 Feature/update docs (#3205)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
* docs: add create_directory parameter

* docs: remove string guardrails to focus on function guardrails

* docs: remove get help from docs.json

* docs: update pt-BR docs.json changes
2025-07-22 13:55:27 -04:00
Lucas Gomide
27623a1d01 feat: remove duplicate print on LLM call error (#3183)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
By improving litellm handler error / outputs

Co-authored-by: Lorenze Jay <63378463+lorenzejay@users.noreply.github.com>
2025-07-21 22:08:07 -04:00
João Moura
2593242234 Adding Support to adhoc tool calling using the internal LLM class (#3195)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
* Adding Support to adhoc tool calling using the internal LLM class

* fix type
2025-07-21 19:36:48 -03:00
Greyson LaLonde
2ab6c31544 chore: add deprecation notices to UserMemory (#3201)
- Mark UserMemory and UserMemoryItem for removal in v0.156.0 or 2025-08-04
- Update all references with deprecation warnings
- Users should migrate to ExternalMemory
2025-07-21 15:26:34 -04:00
Lucas Gomide
3c55c8a22a fix: append user message when last message is from assistent when using Ollama models (#3200)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Ollama doesn't supports last message to be 'assistant'
We can drop this commit after merging https://github.com/BerriAI/litellm/pull/10917
2025-07-21 13:30:40 -04:00
Ranuga Disansa
424433ff58 docs: Add Tavily Search & Extractor tools to Search-Research suite (#3146)
* docs: Add Tavily Search and Extractor tools documentation

* docs: Add Tavily Search and Extractor tools to the documentation

---------

Co-authored-by: Tony Kipkemboi <iamtonykipkemboi@gmail.com>
2025-07-21 12:01:29 -04:00
Lucas Gomide
2fd99503ed build: upgrade LiteLLM to 1.74.3 (#3199) 2025-07-21 09:58:47 -04:00
Vidit Ostwal
942014962e fixed save method, changed the test cases (#3187)
Some checks failed
Notify Downstream / notify-downstream (push) Has been cancelled
Mark stale issues and pull requests / stale (push) Has been cancelled
* fixed save method, changed the test cases

* Linting fixed
2025-07-18 15:10:26 -04:00
Lucas Gomide
2ab79a7dd5 feat: drop unsupported stop parameter for LLM models automatically (#3184) 2025-07-18 13:54:28 -04:00
51 changed files with 5094 additions and 3499 deletions

View File

@@ -32,11 +32,6 @@
"href": "https://chatgpt.com/g/g-qqTuUWsBY-crewai-assistant",
"icon": "robot"
},
{
"anchor": "Get Help",
"href": "mailto:support@crewai.com",
"icon": "headset"
},
{
"anchor": "Releases",
"href": "https://github.com/crewAIInc/crewAI/releases",
@@ -166,7 +161,9 @@
"en/tools/search-research/websitesearchtool",
"en/tools/search-research/codedocssearchtool",
"en/tools/search-research/youtubechannelsearchtool",
"en/tools/search-research/youtubevideosearchtool"
"en/tools/search-research/youtubevideosearchtool",
"en/tools/search-research/tavilysearchtool",
"en/tools/search-research/tavilyextractortool"
]
},
{
@@ -370,11 +367,6 @@
"href": "https://chatgpt.com/g/g-qqTuUWsBY-crewai-assistant",
"icon": "robot"
},
{
"anchor": "Obter Ajuda",
"href": "mailto:support@crewai.com",
"icon": "headset"
},
{
"anchor": "Lançamentos",
"href": "https://github.com/crewAIInc/crewAI/releases",

View File

@@ -270,7 +270,7 @@ In this section, you'll find detailed examples that help you select, configure,
from crewai import LLM
llm = LLM(
model="gemini/gemini-1.5-pro-latest",
model="gemini-1.5-pro-latest", # or vertex_ai/gemini-1.5-pro-latest
temperature=0.7,
vertex_credentials=vertex_credentials_json
)

View File

@@ -623,7 +623,7 @@ for provider in providers_to_test:
**Model not found errors:**
```python
# Verify model availability
from crewai.utilities.embedding_configurator import EmbeddingConfigurator
from crewai.rag.embeddings.configurator import EmbeddingConfigurator
configurator = EmbeddingConfigurator()
try:
@@ -712,7 +712,7 @@ crew = Crew(
memory_config={
"provider": "mem0",
"config": {"user_id": "john"},
"user_memory": {} # Required - triggers user memory initialization
"user_memory": {} # DEPRECATED: Will be removed in version 0.156.0 or on 2025-08-04, use external_memory instead
},
process=Process.sequential,
verbose=True
@@ -720,7 +720,16 @@ crew = Crew(
```
### Advanced Mem0 Configuration
When using Mem0 Client, you can customize the memory configuration further, by using parameters like 'includes', 'excludes', 'custom_categories', 'infer' and 'run_id' (this is only for short-term memory).
You can find more details in the [Mem0 documentation](https://docs.mem0.ai/).
```python
new_categories = [
{"lifestyle_management_concerns": "Tracks daily routines, habits, hobbies and interests including cooking, time management and work-life balance"},
{"seeking_structure": "Documents goals around creating routines, schedules, and organized systems in various life areas"},
{"personal_information": "Basic information about the user including name, preferences, and personality traits"}
]
crew = Crew(
agents=[...],
tasks=[...],
@@ -732,6 +741,11 @@ crew = Crew(
"org_id": "my_org_id", # Optional
"project_id": "my_project_id", # Optional
"api_key": "custom-api-key" # Optional - overrides env var
"run_id": "my_run_id", # Optional - for short-term memory
"includes": "include1", # Optional
"excludes": "exclude1", # Optional
"infer": True # Optional defaults to True
"custom_categories": new_categories # Optional - custom categories for user memory
},
"user_memory": {}
}
@@ -761,7 +775,8 @@ crew = Crew(
"provider": "openai",
"config": {"api_key": "your-api-key", "model": "text-embedding-3-small"}
}
}
},
"infer": True # Optional defaults to True
},
"user_memory": {}
}

View File

@@ -54,10 +54,11 @@ crew = Crew(
| **Markdown** _(optional)_ | `markdown` | `Optional[bool]` | Whether the task should instruct the agent to return the final answer formatted in Markdown. 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. |
| **Create Directory** _(optional)_ | `create_directory` | `Optional[bool]` | Whether to create the directory for output_file if it doesn't exist. Defaults to True. |
| **Output JSON** _(optional)_ | `output_json` | `Optional[Type[BaseModel]]` | A Pydantic model to structure the JSON output. |
| **Output Pydantic** _(optional)_ | `output_pydantic` | `Optional[Type[BaseModel]]` | A Pydantic model for task output. |
| **Callback** _(optional)_ | `callback` | `Optional[Any]` | Function/object to be executed after task completion. |
| **Guardrail** _(optional)_ | `guardrail` | `Optional[Union[Callable, str]]` | Function or string description to validate task output before proceeding to next task. |
| **Guardrail** _(optional)_ | `guardrail` | `Optional[Callable]` | Function to validate task output before proceeding to next task. |
## Creating Tasks
@@ -87,7 +88,6 @@ research_task:
expected_output: >
A list with 10 bullet points of the most relevant information about {topic}
agent: researcher
guardrail: ensure each bullet contains a minimum of 100 words
reporting_task:
description: >
@@ -334,9 +334,7 @@ Task guardrails provide a way to validate and transform task outputs before they
are passed to the next task. This feature helps ensure data quality and provides
feedback to agents when their output doesn't meet specific criteria.
**Guardrails can be defined in two ways:**
1. **Function-based guardrails**: Python functions that implement custom validation logic
2. **String-based guardrails**: Natural language descriptions that are automatically converted to LLM-powered validation
Guardrails are implemented as Python functions that contain custom validation logic, giving you complete control over the validation process and ensuring reliable, deterministic results.
### Function-Based Guardrails
@@ -378,82 +376,7 @@ blog_task = Task(
- On success: it returns a tuple of `(bool, Any)`. For example: `(True, validated_result)`
- On Failure: it returns a tuple of `(bool, str)`. For example: `(False, "Error message explain the failure")`
### String-Based Guardrails
String-based guardrails allow you to describe validation criteria in natural language. When you provide a string instead of a function, CrewAI automatically converts it to an `LLMGuardrail` that uses an AI agent to validate the task output.
#### Using String Guardrails in Python
```python Code
from crewai import Task
# Simple string-based guardrail
blog_task = Task(
description="Write a blog post about AI",
expected_output="A blog post under 200 words",
agent=blog_agent,
guardrail="Ensure the blog post is under 200 words and includes practical examples"
)
# More complex validation criteria
research_task = Task(
description="Research AI trends for 2025",
expected_output="A comprehensive research report",
agent=research_agent,
guardrail="Ensure each finding includes a credible source and is backed by recent data from 2024-2025"
)
```
#### Using String Guardrails in YAML
```yaml
research_task:
description: Research the latest AI developments
expected_output: A list of 10 bullet points about AI
agent: researcher
guardrail: ensure each bullet contains a minimum of 100 words
validation_task:
description: Validate the research findings
expected_output: A validation report
agent: validator
guardrail: confirm all sources are from reputable publications and published within the last 2 years
```
#### How String Guardrails Work
When you provide a string guardrail, CrewAI automatically:
1. Creates an `LLMGuardrail` instance using the string as validation criteria
2. Uses the task's agent LLM to power the validation
3. Creates a temporary validation agent that checks the output against your criteria
4. Returns detailed feedback if validation fails
This approach is ideal when you want to use natural language to describe validation rules without writing custom validation functions.
### LLMGuardrail Class
The `LLMGuardrail` class is the underlying mechanism that powers string-based guardrails. You can also use it directly for more advanced control:
```python Code
from crewai import Task
from crewai.tasks.llm_guardrail import LLMGuardrail
from crewai.llm import LLM
# Create a custom LLMGuardrail with specific LLM
custom_guardrail = LLMGuardrail(
description="Ensure the response contains exactly 5 bullet points with proper citations",
llm=LLM(model="gpt-4o-mini")
)
task = Task(
description="Research AI safety measures",
expected_output="A detailed analysis with bullet points",
agent=research_agent,
guardrail=custom_guardrail
)
```
**Note**: When you use a string guardrail, CrewAI automatically creates an `LLMGuardrail` instance using your task's agent LLM. Using `LLMGuardrail` directly gives you more control over the validation process and LLM selection.
### Error Handling Best Practices
@@ -881,21 +804,87 @@ These validations help in maintaining the consistency and reliability of task ex
## Creating Directories when Saving Files
You can now specify if a task should create directories when saving its output to a file. This is particularly useful for organizing outputs and ensuring that file paths are correctly structured.
The `create_directory` parameter controls whether CrewAI should automatically create directories when saving task outputs to files. This feature is particularly useful for organizing outputs and ensuring that file paths are correctly structured, especially when working with complex project hierarchies.
### Default Behavior
By default, `create_directory=True`, which means CrewAI will automatically create any missing directories in the output file path:
```python Code
# ...
save_output_task = Task(
description='Save the summarized AI news to a file',
expected_output='File saved successfully',
agent=research_agent,
tools=[file_save_tool],
output_file='outputs/ai_news_summary.txt',
create_directory=True
# Default behavior - directories are created automatically
report_task = Task(
description='Generate a comprehensive market analysis report',
expected_output='A detailed market analysis with charts and insights',
agent=analyst_agent,
output_file='reports/2025/market_analysis.md', # Creates 'reports/2025/' if it doesn't exist
markdown=True
)
```
#...
### Disabling Directory Creation
If you want to prevent automatic directory creation and ensure that the directory already exists, set `create_directory=False`:
```python Code
# Strict mode - directory must already exist
strict_output_task = Task(
description='Save critical data that requires existing infrastructure',
expected_output='Data saved to pre-configured location',
agent=data_agent,
output_file='secure/vault/critical_data.json',
create_directory=False # Will raise RuntimeError if 'secure/vault/' doesn't exist
)
```
### YAML Configuration
You can also configure this behavior in your YAML task definitions:
```yaml tasks.yaml
analysis_task:
description: >
Generate quarterly financial analysis
expected_output: >
A comprehensive financial report with quarterly insights
agent: financial_analyst
output_file: reports/quarterly/q4_2024_analysis.pdf
create_directory: true # Automatically create 'reports/quarterly/' directory
audit_task:
description: >
Perform compliance audit and save to existing audit directory
expected_output: >
A compliance audit report
agent: auditor
output_file: audit/compliance_report.md
create_directory: false # Directory must already exist
```
### Use Cases
**Automatic Directory Creation (`create_directory=True`):**
- Development and prototyping environments
- Dynamic report generation with date-based folders
- Automated workflows where directory structure may vary
- Multi-tenant applications with user-specific folders
**Manual Directory Management (`create_directory=False`):**
- Production environments with strict file system controls
- Security-sensitive applications where directories must be pre-configured
- Systems with specific permission requirements
- Compliance environments where directory creation is audited
### Error Handling
When `create_directory=False` and the directory doesn't exist, CrewAI will raise a `RuntimeError`:
```python Code
try:
result = crew.kickoff()
except RuntimeError as e:
# Handle missing directory error
print(f"Directory creation failed: {e}")
# Create directory manually or use fallback location
```
Check out the video below to see how to use structured outputs in CrewAI:

View File

@@ -0,0 +1,356 @@
---
title: LangDB Integration
description: How to use LangDB AI Gateway with CrewAI
icon: database
---
<img src="https://raw.githubusercontent.com/LangDB/assets/main/langdb-crewai-header.png" alt="LangDB CrewAI Header Image" width="70%" />
## Introduction
LangDB is the fastest enterprise AI gateway that enhances CrewAI with production-ready observability and optimization features. It provides:
- **Complete end-to-end tracing** of every agent interaction and LLM call
- **Real-time cost monitoring** and optimization across 250+ LLMs
- **Performance analytics** with detailed metrics and insights
- **Secure governance** for enterprise AI deployments
- **OpenAI-compatible APIs** for seamless integration
- **Fine-grained control** over agent workflows and resource usage
### Installation & Setup
<Steps>
<Step title="Install the required packages">
```bash
pip install -U crewai langdb
```
</Step>
<Step title="Set up environment variables" icon="lock">
Configure your LangDB credentials from the [LangDB dashboard](https://app.langdb.ai/):
```bash
export LANGDB_API_KEY="your_langdb_api_key"
export LANGDB_PROJECT_ID="your_project_id"
```
</Step>
<Step title="Initialize LangDB with CrewAI">
The integration requires a single initialization call before creating your agents:
```python
from langdb import LangDB
from crewai import Agent, Task, Crew, LLM
# Initialize LangDB tracing
LangDB.init()
# Create LLM instance - LangDB automatically traces all calls
llm = LLM(
model="gpt-4o",
temperature=0.7
)
# Create your agents as usual
@agent
def research_agent(self) -> Agent:
return Agent(
role="Senior Research Analyst",
goal="Conduct comprehensive research on assigned topics",
backstory="You are an expert researcher with deep analytical skills.",
llm=llm,
verbose=True
)
```
</Step>
</Steps>
## Key Features
### 1. Comprehensive Observability
LangDB provides complete visibility into your CrewAI agent workflows with minimal setup overhead.
<Tabs>
<Tab title="Request Tracing">
LangDB automatically captures every LLM interaction in your crew execution:
```python
from langdb import LangDB
from crewai import Agent, Task, Crew, LLM
# Initialize with custom trace metadata
LangDB.init(
metadata={
"environment": "production",
"crew_type": "research_workflow",
"user_id": "user_123"
}
)
# All agent interactions are automatically traced
crew = Crew(
agents=[research_agent, writer_agent],
tasks=[research_task, writing_task],
verbose=True
)
# Execute with full tracing
result = crew.kickoff(inputs={"topic": "AI trends 2025"})
```
View detailed traces in the LangDB dashboard showing:
- Complete agent conversation flows
- Tool usage and function calls
- Task execution timelines
- LLM request/response pairs
</Tab>
<Tab title="Performance Metrics">
LangDB tracks comprehensive performance metrics for your crews:
- **Execution Time**: Total and per-task execution duration
- **Token Usage**: Input/output tokens for cost optimization
- **Success Rates**: Task completion and failure analytics
- **Latency Analysis**: Response times and bottleneck identification
```python
# Access metrics programmatically
from langdb import LangDB
# Get crew execution metrics
metrics = LangDB.get_metrics(
project_id="your_project_id",
filters={
"crew_type": "research_workflow",
"time_range": "last_24h"
}
)
print(f"Average execution time: {metrics.avg_execution_time}")
print(f"Total cost: ${metrics.total_cost}")
print(f"Success rate: {metrics.success_rate}%")
```
</Tab>
<Tab title="Cost Monitoring">
Track and optimize AI spending across your CrewAI deployments:
```python
from langdb import LangDB
# Initialize with cost tracking
LangDB.init(
cost_tracking=True,
budget_alerts={
"daily_limit": 100.0, # $100 daily limit
"alert_threshold": 0.8 # Alert at 80% of limit
}
)
# LangDB automatically tracks costs for all LLM calls
crew = Crew(agents=[agent], tasks=[task])
result = crew.kickoff()
# View cost breakdown
cost_report = LangDB.get_cost_report(
breakdown_by=["model", "agent", "task"]
)
```
Features include:
- Real-time cost tracking across all models
- Budget alerts and spending limits
- Cost optimization recommendations
- Detailed cost attribution by agent and task
</Tab>
</Tabs>
### 2. Advanced Analytics & Insights
LangDB provides powerful analytics to optimize your CrewAI workflows.
<Tabs>
<Tab title="Agent Performance Analysis">
Analyze individual agent performance and identify optimization opportunities:
```python
from langdb import LangDB
# Get agent-specific analytics
analytics = LangDB.get_agent_analytics(
agent_role="Senior Research Analyst",
time_range="last_week"
)
print(f"Average task completion time: {analytics.avg_completion_time}")
print(f"Most used tools: {analytics.top_tools}")
print(f"Success rate: {analytics.success_rate}%")
print(f"Cost per task: ${analytics.cost_per_task}")
```
</Tab>
<Tab title="Workflow Optimization">
Identify bottlenecks and optimization opportunities in your crew workflows:
```python
# Analyze crew workflow patterns
workflow_analysis = LangDB.analyze_workflow(
crew_id="research_crew_v1",
optimization_focus=["speed", "cost", "quality"]
)
# Get optimization recommendations
recommendations = workflow_analysis.recommendations
for rec in recommendations:
print(f"Optimization: {rec.type}")
print(f"Potential savings: {rec.estimated_savings}")
print(f"Implementation: {rec.implementation_guide}")
```
</Tab>
</Tabs>
### 3. Production-Ready Features
<CardGroup cols="2">
<Card title="Error Monitoring" icon="exclamation-triangle" href="https://docs.langdb.ai/features/error-monitoring">
Automatic detection and alerting for agent failures, LLM errors, and workflow issues.
</Card>
<Card title="Rate Limiting" icon="gauge" href="https://docs.langdb.ai/features/rate-limiting">
Intelligent rate limiting to prevent API quota exhaustion and optimize throughput.
</Card>
<Card title="Caching" icon="bolt" href="https://docs.langdb.ai/features/caching">
Smart caching of LLM responses to reduce costs and improve response times.
</Card>
<Card title="Load Balancing" icon="scale-balanced" href="https://docs.langdb.ai/features/load-balancing">
Distribute requests across multiple LLM providers for reliability and performance.
</Card>
</CardGroup>
### 4. Enterprise Security & Governance
LangDB provides enterprise-grade security features for production CrewAI deployments:
```python
from langdb import LangDB
# Initialize with security configurations
LangDB.init(
security_config={
"pii_detection": True,
"content_filtering": True,
"audit_logging": True,
"data_retention_days": 90
}
)
# All crew interactions are automatically secured
crew = Crew(agents=[agent], tasks=[task])
result = crew.kickoff()
```
Security features include:
- **PII Detection**: Automatic detection and redaction of sensitive information
- **Content Filtering**: Block inappropriate or harmful content
- **Audit Logging**: Complete audit trails for compliance
- **Data Governance**: Configurable data retention and privacy controls
## Advanced Configuration
### Custom Metadata and Filtering
Add custom metadata to enable powerful filtering and analytics:
```python
from langdb import LangDB
from crewai import Agent, Crew, Task
# Initialize with rich metadata
LangDB.init(
metadata={
"environment": "production",
"team": "research_team",
"version": "v2.1.0",
"customer_tier": "enterprise"
}
)
# Add task-specific metadata
@task
def research_task(self) -> Task:
return Task(
description="Research the latest AI trends",
expected_output="Comprehensive research report",
agent=research_agent,
metadata={
"task_type": "research",
"priority": "high",
"estimated_duration": "30min"
}
)
```
### Multi-Environment Setup
Configure different LangDB projects for different environments:
```python
import os
from langdb import LangDB
# Environment-specific configuration
environment = os.getenv("ENVIRONMENT", "development")
if environment == "production":
LangDB.init(
project_id="prod_project_id",
sampling_rate=1.0, # Trace all requests
cost_tracking=True
)
elif environment == "staging":
LangDB.init(
project_id="staging_project_id",
sampling_rate=0.5, # Sample 50% of requests
cost_tracking=False
)
else:
LangDB.init(
project_id="dev_project_id",
sampling_rate=0.1, # Sample 10% of requests
cost_tracking=False
)
```
## Best Practices
### Development Phase
- Use detailed tracing to understand agent behavior patterns
- Monitor resource usage during testing and development
- Set up cost alerts to prevent unexpected spending
- Implement comprehensive error handling and monitoring
### Production Phase
- Enable full request tracing for complete observability
- Set up automated alerts for performance degradation
- Implement cost optimization strategies based on analytics
- Use metadata for detailed filtering and analysis
### Continuous Improvement
- Regular performance reviews using LangDB analytics
- A/B testing of different agent configurations
- Cost optimization based on usage patterns
- Workflow optimization using bottleneck analysis
## Getting Started
1. **Sign up** for a LangDB account at [app.langdb.ai](https://app.langdb.ai)
2. **Install** the LangDB package: `pip install langdb`
3. **Initialize** LangDB in your CrewAI application
4. **Deploy** your crews with automatic observability
5. **Monitor** and optimize using the LangDB dashboard
<Card title="LangDB Documentation" icon="book" href="https://docs.langdb.ai">
Explore comprehensive LangDB documentation and advanced features
</Card>
LangDB transforms your CrewAI agents into production-ready, observable, and optimized AI workflows with minimal code changes and maximum insights.

View File

@@ -56,6 +56,10 @@ Observability is crucial for understanding how your CrewAI agents perform, ident
<Card title="Weave" icon="network-wired" href="/en/observability/weave">
Weights & Biases platform for tracking and evaluating AI applications.
</Card>
<Card title="LangDB" icon="database" href="/en/observability/langdb">
Enterprise AI gateway with comprehensive tracing, cost optimization, and performance analytics.
</Card>
</CardGroup>
### Evaluation & Quality Assurance

View File

@@ -44,6 +44,14 @@ These tools enable your agents to search the web, research topics, and find info
<Card title="YouTube Video Search" icon="play" href="/en/tools/search-research/youtubevideosearchtool">
Find and analyze YouTube videos by topic, keyword, or criteria.
</Card>
<Card title="Tavily Search Tool" icon="magnifying-glass" href="/en/tools/search-research/tavilysearchtool">
Comprehensive web search using Tavily's AI-powered search API.
</Card>
<Card title="Tavily Extractor Tool" icon="file-text" href="/en/tools/search-research/tavilyextractortool">
Extract structured content from web pages using the Tavily API.
</Card>
</CardGroup>
## **Common Use Cases**
@@ -55,17 +63,19 @@ These tools enable your agents to search the web, research topics, and find info
- **Academic Research**: Find scholarly articles and technical papers
```python
from crewai_tools import SerperDevTool, GitHubSearchTool, YoutubeVideoSearchTool
from crewai_tools import SerperDevTool, GitHubSearchTool, YoutubeVideoSearchTool, TavilySearchTool, TavilyExtractorTool
# Create research tools
web_search = SerperDevTool()
code_search = GitHubSearchTool()
video_research = YoutubeVideoSearchTool()
tavily_search = TavilySearchTool()
content_extractor = TavilyExtractorTool()
# Add to your agent
agent = Agent(
role="Research Analyst",
tools=[web_search, code_search, video_research],
tools=[web_search, code_search, video_research, tavily_search, content_extractor],
goal="Gather comprehensive information on any topic"
)
```

View File

@@ -6,10 +6,6 @@ icon: google
# `SerperDevTool`
<Note>
We are still working on improving tools, so there might be unexpected behavior or changes in the future.
</Note>
## Description
This tool is designed to perform a semantic search for a specified query from a text's content across the internet. It utilizes the [serper.dev](https://serper.dev) API
@@ -17,6 +13,12 @@ to fetch and display the most relevant search results based on the query provide
## Installation
To effectively use the `SerperDevTool`, follow these steps:
1. **Package Installation**: Confirm that the `crewai[tools]` package is installed in your Python environment.
2. **API Key Acquisition**: Acquire a `serper.dev` API key by registering for a free account at `serper.dev`.
3. **Environment Configuration**: Store your obtained API key in an environment variable named `SERPER_API_KEY` to facilitate its use by the tool.
To incorporate this tool into your project, follow the installation instructions below:
```shell
@@ -34,14 +36,6 @@ from crewai_tools import SerperDevTool
tool = SerperDevTool()
```
## Steps to Get Started
To effectively use the `SerperDevTool`, follow these steps:
1. **Package Installation**: Confirm that the `crewai[tools]` package is installed in your Python environment.
2. **API Key Acquisition**: Acquire a `serper.dev` API key by registering for a free account at `serper.dev`.
3. **Environment Configuration**: Store your obtained API key in an environment variable named `SERPER_API_KEY` to facilitate its use by the tool.
## Parameters
The `SerperDevTool` comes with several parameters that will be passed to the API :

View File

@@ -0,0 +1,139 @@
---
title: "Tavily Extractor Tool"
description: "Extract structured content from web pages using the Tavily API"
icon: "file-text"
---
The `TavilyExtractorTool` allows CrewAI agents to extract structured content from web pages using the Tavily API. It can process single URLs or lists of URLs and provides options for controlling the extraction depth and including images.
## Installation
To use the `TavilyExtractorTool`, you need to install the `tavily-python` library:
```shell
pip install 'crewai[tools]' tavily-python
```
You also need to set your Tavily API key as an environment variable:
```bash
export TAVILY_API_KEY='your-tavily-api-key'
```
## Example Usage
Here's how to initialize and use the `TavilyExtractorTool` within a CrewAI agent:
```python
import os
from crewai import Agent, Task, Crew
from crewai_tools import TavilyExtractorTool
# Ensure TAVILY_API_KEY is set in your environment
# os.environ["TAVILY_API_KEY"] = "YOUR_API_KEY"
# Initialize the tool
tavily_tool = TavilyExtractorTool()
# Create an agent that uses the tool
extractor_agent = Agent(
role='Web Content Extractor',
goal='Extract key information from specified web pages',
backstory='You are an expert at extracting relevant content from websites using the Tavily API.',
tools=[tavily_tool],
verbose=True
)
# Define a task for the agent
extract_task = Task(
description='Extract the main content from the URL https://example.com using basic extraction depth.',
expected_output='A JSON string containing the extracted content from the URL.',
agent=extractor_agent
)
# Create and run the crew
crew = Crew(
agents=[extractor_agent],
tasks=[extract_task],
verbose=2
)
result = crew.kickoff()
print(result)
```
## Configuration Options
The `TavilyExtractorTool` accepts the following arguments:
- `urls` (Union[List[str], str]): **Required**. A single URL string or a list of URL strings to extract data from.
- `include_images` (Optional[bool]): Whether to include images in the extraction results. Defaults to `False`.
- `extract_depth` (Literal["basic", "advanced"]): The depth of extraction. Use `"basic"` for faster, surface-level extraction or `"advanced"` for more comprehensive extraction. Defaults to `"basic"`.
- `timeout` (int): The maximum time in seconds to wait for the extraction request to complete. Defaults to `60`.
## Advanced Usage
### Multiple URLs with Advanced Extraction
```python
# Example with multiple URLs and advanced extraction
multi_extract_task = Task(
description='Extract content from https://example.com and https://anotherexample.org using advanced extraction.',
expected_output='A JSON string containing the extracted content from both URLs.',
agent=extractor_agent
)
# Configure the tool with custom parameters
custom_extractor = TavilyExtractorTool(
extract_depth='advanced',
include_images=True,
timeout=120
)
agent_with_custom_tool = Agent(
role="Advanced Content Extractor",
goal="Extract comprehensive content with images",
tools=[custom_extractor]
)
```
### Tool Parameters
You can customize the tool's behavior by setting parameters during initialization:
```python
# Initialize with custom configuration
extractor_tool = TavilyExtractorTool(
extract_depth='advanced', # More comprehensive extraction
include_images=True, # Include image results
timeout=90 # Custom timeout
)
```
## Features
- **Single or Multiple URLs**: Extract content from one URL or process multiple URLs in a single request
- **Configurable Depth**: Choose between basic (fast) and advanced (comprehensive) extraction modes
- **Image Support**: Optionally include images in the extraction results
- **Structured Output**: Returns well-formatted JSON containing the extracted content
- **Error Handling**: Robust handling of network timeouts and extraction errors
## Response Format
The tool returns a JSON string representing the structured data extracted from the provided URL(s). The exact structure depends on the content of the pages and the `extract_depth` used.
Common response elements include:
- **Title**: The page title
- **Content**: Main text content of the page
- **Images**: Image URLs and metadata (when `include_images=True`)
- **Metadata**: Additional page information like author, description, etc.
## Use Cases
- **Content Analysis**: Extract and analyze content from competitor websites
- **Research**: Gather structured data from multiple sources for analysis
- **Content Migration**: Extract content from existing websites for migration
- **Monitoring**: Regular extraction of content for change detection
- **Data Collection**: Systematic extraction of information from web sources
Refer to the [Tavily API documentation](https://docs.tavily.com/docs/tavily-api/python-sdk#extract) for detailed information about the response structure and available options.

View File

@@ -0,0 +1,122 @@
---
title: "Tavily Search Tool"
description: "Perform comprehensive web searches using the Tavily Search API"
icon: "magnifying-glass"
---
The `TavilySearchTool` provides an interface to the Tavily Search API, enabling CrewAI agents to perform comprehensive web searches. It allows for specifying search depth, topics, time ranges, included/excluded domains, and whether to include direct answers, raw content, or images in the results.
## Installation
To use the `TavilySearchTool`, you need to install the `tavily-python` library:
```shell
pip install 'crewai[tools]' tavily-python
```
## Environment Variables
Ensure your Tavily API key is set as an environment variable:
```bash
export TAVILY_API_KEY='your_tavily_api_key'
```
## Example Usage
Here's how to initialize and use the `TavilySearchTool` within a CrewAI agent:
```python
import os
from crewai import Agent, Task, Crew
from crewai_tools import TavilySearchTool
# Ensure the TAVILY_API_KEY environment variable is set
# os.environ["TAVILY_API_KEY"] = "YOUR_TAVILY_API_KEY"
# Initialize the tool
tavily_tool = TavilySearchTool()
# Create an agent that uses the tool
researcher = Agent(
role='Market Researcher',
goal='Find information about the latest AI trends',
backstory='An expert market researcher specializing in technology.',
tools=[tavily_tool],
verbose=True
)
# Create a task for the agent
research_task = Task(
description='Search for the top 3 AI trends in 2024.',
expected_output='A JSON report summarizing the top 3 AI trends found.',
agent=researcher
)
# Form the crew and kick it off
crew = Crew(
agents=[researcher],
tasks=[research_task],
verbose=2
)
result = crew.kickoff()
print(result)
```
## Configuration Options
The `TavilySearchTool` accepts the following arguments during initialization or when calling the `run` method:
- `query` (str): **Required**. The search query string.
- `search_depth` (Literal["basic", "advanced"], optional): The depth of the search. Defaults to `"basic"`.
- `topic` (Literal["general", "news", "finance"], optional): The topic to focus the search on. Defaults to `"general"`.
- `time_range` (Literal["day", "week", "month", "year"], optional): The time range for the search. Defaults to `None`.
- `days` (int, optional): The number of days to search back. Relevant if `time_range` is not set. Defaults to `7`.
- `max_results` (int, optional): The maximum number of search results to return. Defaults to `5`.
- `include_domains` (Sequence[str], optional): A list of domains to prioritize in the search. Defaults to `None`.
- `exclude_domains` (Sequence[str], optional): A list of domains to exclude from the search. Defaults to `None`.
- `include_answer` (Union[bool, Literal["basic", "advanced"]], optional): Whether to include a direct answer synthesized from the search results. Defaults to `False`.
- `include_raw_content` (bool, optional): Whether to include the raw HTML content of the searched pages. Defaults to `False`.
- `include_images` (bool, optional): Whether to include image results. Defaults to `False`.
- `timeout` (int, optional): The request timeout in seconds. Defaults to `60`.
## Advanced Usage
You can configure the tool with custom parameters:
```python
# Example: Initialize with specific parameters
custom_tavily_tool = TavilySearchTool(
search_depth='advanced',
max_results=10,
include_answer=True
)
# The agent will use these defaults
agent_with_custom_tool = Agent(
role="Advanced Researcher",
goal="Conduct detailed research with comprehensive results",
tools=[custom_tavily_tool]
)
```
## Features
- **Comprehensive Search**: Access to Tavily's powerful search index
- **Configurable Depth**: Choose between basic and advanced search modes
- **Topic Filtering**: Focus searches on general, news, or finance topics
- **Time Range Control**: Limit results to specific time periods
- **Domain Control**: Include or exclude specific domains
- **Direct Answers**: Get synthesized answers from search results
- **Content Filtering**: Prevent context window issues with automatic content truncation
## Response Format
The tool returns search results as a JSON string containing:
- Search results with titles, URLs, and content snippets
- Optional direct answers to queries
- Optional image results
- Optional raw HTML content (when enabled)
Content for each result is automatically truncated to prevent context window issues while maintaining the most relevant information.

View File

@@ -0,0 +1,100 @@
---
title: Serper Scrape Website
description: The `SerperScrapeWebsiteTool` is designed to scrape websites and extract clean, readable content using Serper's scraping API.
icon: globe
---
# `SerperScrapeWebsiteTool`
## Description
This tool is designed to scrape website content and extract clean, readable text from any website URL. It utilizes the [serper.dev](https://serper.dev) scraping API to fetch and process web pages, optionally including markdown formatting for better structure and readability.
## Installation
To effectively use the `SerperScrapeWebsiteTool`, follow these steps:
1. **Package Installation**: Confirm that the `crewai[tools]` package is installed in your Python environment.
2. **API Key Acquisition**: Acquire a `serper.dev` API key by registering for an account at `serper.dev`.
3. **Environment Configuration**: Store your obtained API key in an environment variable named `SERPER_API_KEY` to facilitate its use by the tool.
To incorporate this tool into your project, follow the installation instructions below:
```shell
pip install 'crewai[tools]'
```
## Example
The following example demonstrates how to initialize the tool and scrape a website:
```python Code
from crewai_tools import SerperScrapeWebsiteTool
# Initialize the tool for website scraping capabilities
tool = SerperScrapeWebsiteTool()
# Scrape a website with markdown formatting
result = tool.run(url="https://example.com", include_markdown=True)
```
## Arguments
The `SerperScrapeWebsiteTool` accepts the following arguments:
- **url**: Required. The URL of the website to scrape.
- **include_markdown**: Optional. Whether to include markdown formatting in the scraped content. Defaults to `True`.
## Example with Parameters
Here is an example demonstrating how to use the tool with different parameters:
```python Code
from crewai_tools import SerperScrapeWebsiteTool
tool = SerperScrapeWebsiteTool()
# Scrape with markdown formatting (default)
markdown_result = tool.run(
url="https://docs.crewai.com",
include_markdown=True
)
# Scrape without markdown formatting for plain text
plain_result = tool.run(
url="https://docs.crewai.com",
include_markdown=False
)
print("Markdown formatted content:")
print(markdown_result)
print("\nPlain text content:")
print(plain_result)
```
## Use Cases
The `SerperScrapeWebsiteTool` is particularly useful for:
- **Content Analysis**: Extract and analyze website content for research purposes
- **Data Collection**: Gather structured information from web pages
- **Documentation Processing**: Convert web-based documentation into readable formats
- **Competitive Analysis**: Scrape competitor websites for market research
- **Content Migration**: Extract content from existing websites for migration purposes
## Error Handling
The tool includes comprehensive error handling for:
- **Network Issues**: Handles connection timeouts and network errors gracefully
- **API Errors**: Provides detailed error messages for API-related issues
- **Invalid URLs**: Validates and reports issues with malformed URLs
- **Authentication**: Clear error messages for missing or invalid API keys
## Security Considerations
- Always store your `SERPER_API_KEY` in environment variables, never hardcode it in your source code
- Be mindful of rate limits imposed by the Serper API
- Respect robots.txt and website terms of service when scraping content
- Consider implementing delays between requests for large-scale scraping operations

View File

@@ -84,8 +84,8 @@ filename = "seu_modelo.pkl"
try:
SuaCrew().crew().train(
n_iterations=n_iterations,
inputs=inputs,
n_iterations=n_iterations,
inputs=inputs,
filename=filename
)
except Exception as e:
@@ -103,7 +103,7 @@ crewai replay [OPTIONS]
- `-t, --task_id TEXT`: Reexecuta o crew a partir deste task ID, incluindo todas as tarefas subsequentes
Exemplo:
```shell Terminal
```shell Terminal
crewai replay -t task_123456
```
@@ -149,7 +149,7 @@ crewai test [OPTIONS]
- `-m, --model TEXT`: Modelo LLM para executar os testes no Crew (padrão: "gpt-4o-mini")
Exemplo:
```shell Terminal
```shell Terminal
crewai test -n 5 -m gpt-3.5-turbo
```
@@ -203,10 +203,7 @@ def crew(self) -> Crew:
Implemente o crew ou flow no [CrewAI Enterprise](https://app.crewai.com).
- **Autenticação**: Você precisa estar autenticado para implementar no CrewAI Enterprise.
```shell Terminal
crewai signup
```
Caso já tenha uma conta, você pode fazer login com:
Você pode fazer login ou criar uma conta com:
```shell Terminal
crewai login
```
@@ -253,7 +250,7 @@ Você deve estar autenticado no CrewAI Enterprise para usar estes comandos de ge
- **Implantar o Crew**: Depois de autenticado, você pode implantar seu crew ou flow no CrewAI Enterprise.
```shell Terminal
crewai deploy push
```
```
- Inicia o processo de deployment na plataforma CrewAI Enterprise.
- Após a iniciação bem-sucedida, será exibida a mensagem Deployment created successfully! juntamente com o Nome do Deployment e um Deployment ID (UUID) único.
@@ -326,4 +323,4 @@ Ao escolher um provedor, o CLI solicitará que você informe o nome da chave e a
Veja o seguinte link para o nome de chave de cada provedor:
* [LiteLLM Providers](https://docs.litellm.ai/docs/providers)
* [LiteLLM Providers](https://docs.litellm.ai/docs/providers)

View File

@@ -268,7 +268,7 @@ Nesta seção, você encontrará exemplos detalhados que ajudam a selecionar, co
from crewai import LLM
llm = LLM(
model="gemini/gemini-1.5-pro-latest",
model="gemini-1.5-pro-latest", # or vertex_ai/gemini-1.5-pro-latest
temperature=0.7,
vertex_credentials=vertex_credentials_json
)

View File

@@ -623,7 +623,7 @@ for provider in providers_to_test:
**Erros de modelo não encontrado:**
```python
# Verifique disponibilidade do modelo
from crewai.utilities.embedding_configurator import EmbeddingConfigurator
from crewai.rag.embeddings.configurator import EmbeddingConfigurator
configurator = EmbeddingConfigurator()
try:

View File

@@ -54,10 +54,11 @@ crew = Crew(
| **Markdown** _(opcional)_ | `markdown` | `Optional[bool]` | Se a tarefa deve instruir o agente a retornar a resposta final formatada em Markdown. O padrão é False. |
| **Config** _(opcional)_ | `config` | `Optional[Dict[str, Any]]` | Parâmetros de configuração específicos da tarefa. |
| **Arquivo de Saída** _(opcional)_| `output_file` | `Optional[str]` | Caminho do arquivo para armazenar a saída da tarefa. |
| **Criar Diretório** _(opcional)_ | `create_directory` | `Optional[bool]` | Se deve criar o diretório para output_file caso não exista. O padrão é True. |
| **Saída JSON** _(opcional)_ | `output_json` | `Optional[Type[BaseModel]]` | Um modelo Pydantic para estruturar a saída em JSON. |
| **Output Pydantic** _(opcional)_ | `output_pydantic` | `Optional[Type[BaseModel]]` | Um modelo Pydantic para a saída da tarefa. |
| **Callback** _(opcional)_ | `callback` | `Optional[Any]` | Função/objeto a ser executado após a conclusão da tarefa. |
| **Guardrail** _(opcional)_ | `guardrail` | `Optional[Union[Callable, str]]` | Função ou descrição em string para validar a saída da tarefa antes de prosseguir para a próxima tarefa. |
| **Guardrail** _(opcional)_ | `guardrail` | `Optional[Callable]` | Função para validar a saída da tarefa antes de prosseguir para a próxima tarefa. |
## Criando Tarefas
@@ -87,7 +88,6 @@ research_task:
expected_output: >
Uma lista com 10 tópicos em bullet points das informações mais relevantes sobre {topic}
agent: researcher
guardrail: garanta que cada bullet point contenha no mínimo 100 palavras
reporting_task:
description: >
@@ -332,9 +332,7 @@ analysis_task = Task(
Guardrails (trilhas de proteção) de tarefas fornecem uma maneira de validar e transformar as saídas das tarefas antes que elas sejam passadas para a próxima tarefa. Esse recurso assegura a qualidade dos dados e oferece feedback aos agentes quando sua saída não atende a critérios específicos.
**Guardrails podem ser definidos de duas maneiras:**
1. **Guardrails baseados em função**: Funções Python que implementam lógica de validação customizada
2. **Guardrails baseados em string**: Descrições em linguagem natural que são automaticamente convertidas em validação baseada em LLM
Guardrails são implementados como funções Python que contêm lógica de validação customizada, proporcionando controle total sobre o processo de validação e garantindo resultados confiáveis e determinísticos.
### Guardrails Baseados em Função
@@ -376,82 +374,7 @@ blog_task = Task(
- Em caso de sucesso: retorna uma tupla `(True, resultado_validado)`
- Em caso de falha: retorna uma tupla `(False, "mensagem de erro explicando a falha")`
### Guardrails Baseados em String
Guardrails baseados em string permitem que você descreva critérios de validação em linguagem natural. Quando você fornece uma string em vez de uma função, o CrewAI automaticamente a converte em um `LLMGuardrail` que usa um agente de IA para validar a saída da tarefa.
#### Usando Guardrails de String em Python
```python Code
from crewai import Task
# Guardrail simples baseado em string
blog_task = Task(
description="Escreva um post de blog sobre IA",
expected_output="Um post de blog com menos de 200 palavras",
agent=blog_agent,
guardrail="Garanta que o post do blog tenha menos de 200 palavras e inclua exemplos práticos"
)
# Critérios de validação mais complexos
research_task = Task(
description="Pesquise tendências de IA para 2025",
expected_output="Um relatório abrangente de pesquisa",
agent=research_agent,
guardrail="Garanta que cada descoberta inclua uma fonte confiável e seja respaldada por dados recentes de 2024-2025"
)
```
#### Usando Guardrails de String em YAML
```yaml
research_task:
description: Pesquise os últimos desenvolvimentos em IA
expected_output: Uma lista de 10 bullet points sobre IA
agent: researcher
guardrail: garanta que cada bullet point contenha no mínimo 100 palavras
validation_task:
description: Valide os achados da pesquisa
expected_output: Um relatório de validação
agent: validator
guardrail: confirme que todas as fontes são de publicações respeitáveis e publicadas nos últimos 2 anos
```
#### Como Funcionam os Guardrails de String
Quando você fornece um guardrail de string, o CrewAI automaticamente:
1. Cria uma instância `LLMGuardrail` usando a string como critério de validação
2. Usa o LLM do agente da tarefa para alimentar a validação
3. Cria um agente temporário de validação que verifica a saída contra seus critérios
4. Retorna feedback detalhado se a validação falhar
Esta abordagem é ideal quando você quer usar linguagem natural para descrever regras de validação sem escrever funções de validação customizadas.
### Classe LLMGuardrail
A classe `LLMGuardrail` é o mecanismo subjacente que alimenta os guardrails baseados em string. Você também pode usá-la diretamente para maior controle avançado:
```python Code
from crewai import Task
from crewai.tasks.llm_guardrail import LLMGuardrail
from crewai.llm import LLM
# Crie um LLMGuardrail customizado com LLM específico
custom_guardrail = LLMGuardrail(
description="Garanta que a resposta contenha exatamente 5 bullet points com citações adequadas",
llm=LLM(model="gpt-4o-mini")
)
task = Task(
description="Pesquise medidas de segurança em IA",
expected_output="Uma análise detalhada com bullet points",
agent=research_agent,
guardrail=custom_guardrail
)
```
**Nota**: Quando você usa um guardrail de string, o CrewAI automaticamente cria uma instância `LLMGuardrail` usando o LLM do agente da sua tarefa. Usar `LLMGuardrail` diretamente lhe dá mais controle sobre o processo de validação e seleção de LLM.
### Melhores Práticas de Tratamento de Erros
@@ -902,26 +825,7 @@ task = Task(
)
```
#### Use uma abordagem no-code para validação
```python Code
from crewai import Task
task = Task(
description="Gerar dados em JSON",
expected_output="Objeto JSON válido",
guardrail="Garanta que a resposta é um objeto JSON válido"
)
```
#### Usando YAML
```yaml
research_task:
...
guardrail: garanta que cada bullet tenha no mínimo 100 palavras
...
```
```python Code
@CrewBase
@@ -1037,21 +941,87 @@ task = Task(
## Criando Diretórios ao Salvar Arquivos
Agora é possível especificar se uma tarefa deve criar diretórios ao salvar sua saída em arquivo. Isso é útil para organizar outputs e garantir que os caminhos estejam corretos.
O parâmetro `create_directory` controla se o CrewAI deve criar automaticamente diretórios ao salvar saídas de tarefas em arquivos. Este recurso é particularmente útil para organizar outputs e garantir que os caminhos de arquivos estejam estruturados corretamente, especialmente ao trabalhar com hierarquias de projetos complexas.
### Comportamento Padrão
Por padrão, `create_directory=True`, o que significa que o CrewAI criará automaticamente qualquer diretório ausente no caminho do arquivo de saída:
```python Code
# ...
save_output_task = Task(
description='Salve o resumo das notícias de IA em um arquivo',
expected_output='Arquivo salvo com sucesso',
agent=research_agent,
tools=[file_save_tool],
output_file='outputs/ai_news_summary.txt',
create_directory=True
# Comportamento padrão - diretórios são criados automaticamente
report_task = Task(
description='Gerar um relatório abrangente de análise de mercado',
expected_output='Uma análise detalhada de mercado com gráficos e insights',
agent=analyst_agent,
output_file='reports/2025/market_analysis.md', # Cria 'reports/2025/' se não existir
markdown=True
)
```
#...
### Desabilitando a Criação de Diretórios
Se você quiser evitar a criação automática de diretórios e garantir que o diretório já exista, defina `create_directory=False`:
```python Code
# Modo estrito - o diretório já deve existir
strict_output_task = Task(
description='Salvar dados críticos que requerem infraestrutura existente',
expected_output='Dados salvos em localização pré-configurada',
agent=data_agent,
output_file='secure/vault/critical_data.json',
create_directory=False # Gerará RuntimeError se 'secure/vault/' não existir
)
```
### Configuração YAML
Você também pode configurar este comportamento em suas definições de tarefas YAML:
```yaml tasks.yaml
analysis_task:
description: >
Gerar análise financeira trimestral
expected_output: >
Um relatório financeiro abrangente com insights trimestrais
agent: financial_analyst
output_file: reports/quarterly/q4_2024_analysis.pdf
create_directory: true # Criar automaticamente o diretório 'reports/quarterly/'
audit_task:
description: >
Realizar auditoria de conformidade e salvar no diretório de auditoria existente
expected_output: >
Um relatório de auditoria de conformidade
agent: auditor
output_file: audit/compliance_report.md
create_directory: false # O diretório já deve existir
```
### Casos de Uso
**Criação Automática de Diretórios (`create_directory=True`):**
- Ambientes de desenvolvimento e prototipagem
- Geração dinâmica de relatórios com pastas baseadas em datas
- Fluxos de trabalho automatizados onde a estrutura de diretórios pode variar
- Aplicações multi-tenant com pastas específicas do usuário
**Gerenciamento Manual de Diretórios (`create_directory=False`):**
- Ambientes de produção com controles rígidos do sistema de arquivos
- Aplicações sensíveis à segurança onde diretórios devem ser pré-configurados
- Sistemas com requisitos específicos de permissão
- Ambientes de conformidade onde a criação de diretórios é auditada
### Tratamento de Erros
Quando `create_directory=False` e o diretório não existe, o CrewAI gerará um `RuntimeError`:
```python Code
try:
result = crew.kickoff()
except RuntimeError as e:
# Tratar erro de diretório ausente
print(f"Falha na criação do diretório: {e}")
# Criar diretório manualmente ou usar local alternativo
```
Veja o vídeo abaixo para aprender como utilizar saídas estruturadas no CrewAI:

View File

@@ -11,7 +11,7 @@ dependencies = [
# Core Dependencies
"pydantic>=2.4.2",
"openai>=1.13.3",
"litellm==1.72.6",
"litellm==1.74.3",
"instructor>=1.3.3",
# Text Processing
"pdfplumber>=0.11.4",
@@ -48,7 +48,7 @@ Documentation = "https://docs.crewai.com"
Repository = "https://github.com/crewAIInc/crewAI"
[project.optional-dependencies]
tools = ["crewai-tools~=0.55.0"]
tools = ["crewai-tools~=0.58.0"]
embeddings = [
"tiktoken~=0.8.0"
]

View File

@@ -54,7 +54,7 @@ def _track_install_async():
_track_install_async()
__version__ = "0.148.0"
__version__ = "0.150.0"
__all__ = [
"Agent",
"Crew",

View File

@@ -120,11 +120,8 @@ class CrewAgentExecutor(CrewAgentExecutorMixin):
raise
except Exception as e:
handle_unknown_error(self._printer, e)
if e.__class__.__module__.startswith("litellm"):
# Do not retry on litellm errors
raise e
else:
raise e
raise
if self.ask_for_human_input:
formatted_answer = self._handle_human_feedback(formatted_answer)

View File

@@ -26,7 +26,7 @@ class PlusAPIMixin:
"Please sign up/login to CrewAI+ before using the CLI.",
style="bold red",
)
console.print("Run 'crewai signup' to sign up/login.", style="bold green")
console.print("Run 'crewai login' to sign up/login.", style="bold green")
raise SystemExit
def _validate_response(self, response: requests.Response) -> None:

View File

@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
authors = [{ name = "Your Name", email = "you@example.com" }]
requires-python = ">=3.10,<3.14"
dependencies = [
"crewai[tools]>=0.148.0,<1.0.0"
"crewai[tools]>=0.150.0,<1.0.0"
]
[project.scripts]

View File

@@ -5,7 +5,7 @@ description = "{{name}} using crewAI"
authors = [{ name = "Your Name", email = "you@example.com" }]
requires-python = ">=3.10,<3.14"
dependencies = [
"crewai[tools]>=0.148.0,<1.0.0",
"crewai[tools]>=0.150.0,<1.0.0",
]
[project.scripts]

View File

@@ -5,7 +5,7 @@ description = "Power up your crews with {{folder_name}}"
readme = "README.md"
requires-python = ">=3.10,<3.14"
dependencies = [
"crewai[tools]>=0.148.0"
"crewai[tools]>=0.150.0"
]
[tool.crewai]

View File

@@ -161,7 +161,7 @@ class Crew(FlowTrackable, BaseModel):
)
user_memory: Optional[InstanceOf[UserMemory]] = Field(
default=None,
description="An instance of the UserMemory to be used by the Crew to store/fetch memories of a specific user.",
description="DEPRECATED: Will be removed in version 0.156.0 or on 2025-08-04, whichever comes first. Use external_memory instead.",
)
external_memory: Optional[InstanceOf[ExternalMemory]] = Field(
default=None,
@@ -327,7 +327,7 @@ class Crew(FlowTrackable, BaseModel):
self._short_term_memory = self.short_term_memory
self._entity_memory = self.entity_memory
# UserMemory is gonna to be deprecated in the future, but we have to initialize a default value for now
# UserMemory will be removed in version 0.156.0 or on 2025-08-04, whichever comes first
self._user_memory = None
if self.memory:
@@ -1255,6 +1255,7 @@ class Crew(FlowTrackable, BaseModel):
if self.external_memory:
copied_data["external_memory"] = self.external_memory.model_copy(deep=True)
if self.user_memory:
# DEPRECATED: UserMemory will be removed in version 0.156.0 or on 2025-08-04
copied_data["user_memory"] = self.user_memory.model_copy(deep=True)
copied_data.pop("agents", None)

View File

@@ -436,6 +436,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
_routers: Set[str] = set()
_router_paths: Dict[str, List[str]] = {}
initial_state: Union[Type[T], T, None] = None
name: Optional[str] = None
def __class_getitem__(cls: Type["Flow"], item: Type[T]) -> Type["Flow"]:
class _FlowGeneric(cls): # type: ignore
@@ -473,7 +474,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
self,
FlowCreatedEvent(
type="flow_created",
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
),
)
@@ -769,7 +770,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
self,
FlowStartedEvent(
type="flow_started",
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
inputs=inputs,
),
)
@@ -792,7 +793,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
self,
FlowFinishedEvent(
type="flow_finished",
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
result=final_output,
),
)
@@ -834,7 +835,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
MethodExecutionStartedEvent(
type="method_execution_started",
method_name=method_name,
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
params=dumped_params,
state=self._copy_state(),
),
@@ -856,7 +857,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
MethodExecutionFinishedEvent(
type="method_execution_finished",
method_name=method_name,
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
state=self._copy_state(),
result=result,
),
@@ -869,7 +870,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
MethodExecutionFailedEvent(
type="method_execution_failed",
method_name=method_name,
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
error=e,
),
)
@@ -1076,7 +1077,7 @@ class Flow(Generic[T], metaclass=FlowMeta):
self,
FlowPlotEvent(
type="flow_plot",
flow_name=self.__class__.__name__,
flow_name=self.name or self.__class__.__name__,
),
)
plot_flow(self, filename)

View File

@@ -1,55 +0,0 @@
from abc import ABC, abstractmethod
from typing import List
import numpy as np
class BaseEmbedder(ABC):
"""
Abstract base class for text embedding models
"""
@abstractmethod
def embed_chunks(self, chunks: List[str]) -> np.ndarray:
"""
Generate embeddings for a list of text chunks
Args:
chunks: List of text chunks to embed
Returns:
Array of embeddings
"""
pass
@abstractmethod
def embed_texts(self, texts: List[str]) -> np.ndarray:
"""
Generate embeddings for a list of texts
Args:
texts: List of texts to embed
Returns:
Array of embeddings
"""
pass
@abstractmethod
def embed_text(self, text: str) -> np.ndarray:
"""
Generate embedding for a single text
Args:
text: Text to embed
Returns:
Embedding array
"""
pass
@property
@abstractmethod
def dimension(self) -> int:
"""Get the dimension of the embeddings"""
pass

View File

@@ -13,7 +13,7 @@ from chromadb.api.types import OneOrMany
from chromadb.config import Settings
from crewai.knowledge.storage.base_knowledge_storage import BaseKnowledgeStorage
from crewai.utilities import EmbeddingConfigurator
from crewai.rag.embeddings.configurator import EmbeddingConfigurator
from crewai.utilities.chromadb import sanitize_collection_name
from crewai.utilities.constants import KNOWLEDGE_DIRECTORY
from crewai.utilities.logger import Logger

View File

@@ -59,6 +59,7 @@ from crewai.utilities.exceptions.context_window_exceeding_exception import (
load_dotenv()
litellm.suppress_debug_info = True
class FilteredStream(io.TextIOBase):
_lock = None
@@ -76,9 +77,7 @@ class FilteredStream(io.TextIOBase):
# Skip common noisy LiteLLM banners and any other lines that contain "litellm"
if (
"give feedback / get help" in lower_s
or "litellm.info:" in lower_s
or "litellm" in lower_s
"litellm.info:" in lower_s
or "Consider using a smaller input or implementing a text splitting strategy" in lower_s
):
return 0
@@ -760,7 +759,7 @@ class LLM(BaseLLM):
available_functions: Optional[Dict[str, Any]] = None,
from_task: Optional[Any] = None,
from_agent: Optional[Any] = None,
) -> str:
) -> str | Any:
"""Handle a non-streaming response from the LLM.
Args:
@@ -784,13 +783,11 @@ class LLM(BaseLLM):
# Convert litellm's context window error to our own exception type
# for consistent handling in the rest of the codebase
raise LLMContextLengthExceededException(str(e))
# --- 2) Extract response message and content
response_message = cast(Choices, cast(ModelResponse, response).choices)[
0
].message
text_response = response_message.content or ""
# --- 3) Handle callbacks with usage info
if callbacks and len(callbacks) > 0:
for callback in callbacks:
@@ -803,21 +800,22 @@ class LLM(BaseLLM):
start_time=0,
end_time=0,
)
# --- 4) Check for tool calls
tool_calls = getattr(response_message, "tool_calls", [])
# --- 5) If no tool calls or no available functions, return the text response directly
if not tool_calls or not available_functions:
# --- 5) If no tool calls or no available functions, return the text response directly as long as there is a text response
if (not tool_calls or not available_functions) and text_response:
self._handle_emit_call_events(response=text_response, call_type=LLMCallType.LLM_CALL, from_task=from_task, from_agent=from_agent, messages=params["messages"])
return text_response
# --- 6) If there is no text response, no available functions, but there are tool calls, return the tool calls
elif tool_calls and not available_functions and not text_response:
return tool_calls
# --- 6) Handle tool calls if present
# --- 7) Handle tool calls if present
tool_result = self._handle_tool_call(tool_calls, available_functions)
if tool_result is not None:
return tool_result
# --- 7) If tool call handling didn't return a result, emit completion event and return text response
# --- 8) If tool call handling didn't return a result, emit completion event and return text response
self._handle_emit_call_events(response=text_response, call_type=LLMCallType.LLM_CALL, from_task=from_task, from_agent=from_agent, messages=params["messages"])
return text_response
@@ -952,22 +950,18 @@ class LLM(BaseLLM):
# --- 3) Convert string messages to proper format if needed
if isinstance(messages, str):
messages = [{"role": "user", "content": messages}]
# --- 4) Handle O1 model special case (system messages not supported)
if "o1" in self.model.lower():
for message in messages:
if message.get("role") == "system":
message["role"] = "assistant"
# --- 5) Set up callbacks if provided
with suppress_warnings():
if callbacks and len(callbacks) > 0:
self.set_callbacks(callbacks)
try:
# --- 6) Prepare parameters for the completion call
params = self._prepare_completion_params(messages, tools)
# --- 7) Make the completion call and handle response
if self.stream:
return self._handle_streaming_response(
@@ -984,12 +978,32 @@ class LLM(BaseLLM):
# whether to summarize the content or abort based on the respect_context_window flag
raise
except Exception as e:
unsupported_stop = "Unsupported parameter" in str(e) and "'stop'" in str(e)
if unsupported_stop:
if "additional_drop_params" in self.additional_params and isinstance(self.additional_params["additional_drop_params"], list):
self.additional_params["additional_drop_params"].append("stop")
else:
self.additional_params = {"additional_drop_params": ["stop"]}
logging.info(
"Retrying LLM call without the unsupported 'stop'"
)
return self.call(
messages,
tools=tools,
callbacks=callbacks,
available_functions=available_functions,
from_task=from_task,
from_agent=from_agent,
)
assert hasattr(crewai_event_bus, "emit")
crewai_event_bus.emit(
self,
event=LLMCallFailedEvent(error=str(e), from_task=from_task, from_agent=from_agent),
)
logging.error(f"LiteLLM call failed: {str(e)}")
raise
def _handle_emit_call_events(self, response: Any, call_type: LLMCallType, from_task: Optional[Any] = None, from_agent: Optional[Any] = None, messages: str | list[dict[str, Any]] | None = None):
@@ -1058,6 +1072,15 @@ class LLM(BaseLLM):
messages.append({"role": "user", "content": "Please continue."})
return messages
# TODO: Remove this code after merging PR https://github.com/BerriAI/litellm/pull/10917
# Ollama doesn't supports last message to be 'assistant'
if "ollama" in self.model.lower() and messages and messages[-1]["role"] == "assistant":
messages = messages.copy()
messages.append(
{"role": "user", "content": ""}
)
return messages
# Handle Anthropic models
if not self.is_anthropic:
return messages

View File

@@ -108,6 +108,7 @@ class ContextualMemory:
def _fetch_user_context(self, query: str) -> str:
"""
DEPRECATED: Will be removed in version 0.156.0 or on 2025-08-04, whichever comes first.
Fetches and formats relevant user information from User Memory.
Args:
query (str): The search query to find relevant user memories.

View File

@@ -4,7 +4,6 @@ from typing import Any, Dict, List
from mem0 import Memory, MemoryClient
from crewai.memory.storage.interface import Storage
from crewai.utilities.chromadb import sanitize_collection_name
MAX_AGENT_ID_LENGTH_MEM0 = 255
@@ -13,135 +12,150 @@ class Mem0Storage(Storage):
"""
Extends Storage to handle embedding and searching across entities using Mem0.
"""
def __init__(self, type, crew=None, config=None):
super().__init__()
supported_types = ["user", "short_term", "long_term", "entities", "external"]
if type not in supported_types:
raise ValueError(
f"Invalid type '{type}' for Mem0Storage. Must be one of: "
+ ", ".join(supported_types)
)
self._validate_type(type)
self.memory_type = type
self.crew = crew
self.config = config or {}
# TODO: Memory config will be removed in the future the config will be passed as a parameter
self.memory_config = self.config or getattr(crew, "memory_config", {}) or {}
# User ID is required for user memory type "user" since it's used as a unique identifier for the user.
user_id = self._get_user_id()
if type == "user" and not user_id:
# TODO: Memory config will be removed in the future the config will be passed as a parameter
self.config = config or getattr(crew, "memory_config", {}).get("config", {}) or {}
self._validate_user_id()
self._extract_config_values()
self._initialize_memory()
def _validate_type(self, type):
supported_types = {"user", "short_term", "long_term", "entities", "external"}
if type not in supported_types:
raise ValueError(
f"Invalid type '{type}' for Mem0Storage. Must be one of: {', '.join(supported_types)}"
)
def _validate_user_id(self):
if self.memory_type == "user" and not self.config.get("user_id", ""):
raise ValueError("User ID is required for user memory type")
# API key in memory config overrides the environment variable
config = self._get_config()
mem0_api_key = config.get("api_key") or os.getenv("MEM0_API_KEY")
mem0_org_id = config.get("org_id")
mem0_project_id = config.get("project_id")
mem0_local_config = config.get("local_mem0_config")
def _extract_config_values(self):
cfg = self.config
self.mem0_run_id = cfg.get("run_id")
self.includes = cfg.get("includes")
self.excludes = cfg.get("excludes")
self.custom_categories = cfg.get("custom_categories")
self.infer = cfg.get("infer", True)
# Initialize MemoryClient or Memory based on the presence of the mem0_api_key
if mem0_api_key:
if mem0_org_id and mem0_project_id:
self.memory = MemoryClient(
api_key=mem0_api_key, org_id=mem0_org_id, project_id=mem0_project_id
)
else:
self.memory = MemoryClient(api_key=mem0_api_key)
def _initialize_memory(self):
api_key = self.config.get("api_key") or os.getenv("MEM0_API_KEY")
org_id = self.config.get("org_id")
project_id = self.config.get("project_id")
local_config = self.config.get("local_mem0_config")
if api_key:
self.memory = (
MemoryClient(api_key=api_key, org_id=org_id, project_id=project_id)
if org_id and project_id
else MemoryClient(api_key=api_key)
)
if self.custom_categories:
self.memory.update_project(custom_categories=self.custom_categories)
else:
if mem0_local_config and len(mem0_local_config):
self.memory = Memory.from_config(mem0_local_config)
else:
self.memory = Memory()
self.memory = (
Memory.from_config(local_config)
if local_config and len(local_config)
else Memory()
)
def _sanitize_role(self, role: str) -> str:
def _create_filter_for_search(self):
"""
Sanitizes agent roles to ensure valid directory names.
Returns:
dict: A filter dictionary containing AND conditions for querying data.
- Includes user_id if memory_type is 'external'.
- Includes run_id if memory_type is 'short_term' and mem0_run_id is present.
"""
return role.replace("\n", "").replace(" ", "_").replace("/", "_")
filter = {
"AND": []
}
# Add user_id condition if the memory type is external
if self.memory_type == "external":
filter["AND"].append({"user_id": self.config.get("user_id", "")})
# Add run_id condition if the memory type is short_term and a run ID is set
if self.memory_type == "short_term" and self.mem0_run_id:
filter["AND"].append({"run_id": self.mem0_run_id})
return filter
def save(self, value: Any, metadata: Dict[str, Any]) -> None:
user_id = self._get_user_id()
agent_name = self._get_agent_name()
params = None
if self.memory_type == "short_term":
params = {
"agent_id": agent_name,
"infer": False,
"metadata": {"type": "short_term", **metadata},
}
elif self.memory_type == "long_term":
params = {
"agent_id": agent_name,
"infer": False,
"metadata": {"type": "long_term", **metadata},
}
elif self.memory_type == "entities":
params = {
"agent_id": agent_name,
"infer": False,
"metadata": {"type": "entity", **metadata},
}
elif self.memory_type == "external":
params = {
"user_id": user_id,
"agent_id": agent_name,
"metadata": {"type": "external", **metadata},
}
user_id = self.config.get("user_id", "")
assistant_message = [{"role" : "assistant","content" : value}]
if params:
if isinstance(self.memory, MemoryClient):
params["output_format"] = "v1.1"
self.memory.add(value, **params)
base_metadata = {
"short_term": "short_term",
"long_term": "long_term",
"entities": "entity",
"external": "external"
}
def search(
self,
query: str,
limit: int = 3,
score_threshold: float = 0.35,
) -> List[Any]:
params = {"query": query, "limit": limit, "output_format": "v1.1"}
if user_id := self._get_user_id():
# Shared base params
params: dict[str, Any] = {
"metadata": {"type": base_metadata[self.memory_type], **metadata},
"infer": self.infer
}
if self.memory_type == "external":
params["user_id"] = user_id
agent_name = self._get_agent_name()
if self.memory_type == "short_term":
params["agent_id"] = agent_name
params["metadata"] = {"type": "short_term"}
elif self.memory_type == "long_term":
params["agent_id"] = agent_name
params["metadata"] = {"type": "long_term"}
elif self.memory_type == "entities":
params["agent_id"] = agent_name
params["metadata"] = {"type": "entity"}
elif self.memory_type == "external":
params["agent_id"] = agent_name
params["metadata"] = {"type": "external"}
if params:
# MemoryClient-specific overrides
if isinstance(self.memory, MemoryClient):
params["includes"] = self.includes
params["excludes"] = self.excludes
params["output_format"] = "v1.1"
params["version"]="v2"
if self.memory_type == "short_term":
params["run_id"] = self.mem0_run_id
self.memory.add(assistant_message, **params)
def search(self,query: str,limit: int = 3,score_threshold: float = 0.35) -> List[Any]:
params = {
"query": query,
"limit": limit,
"version": "v2",
"output_format": "v1.1"
}
if user_id := self.config.get("user_id", ""):
params["user_id"] = user_id
memory_type_map = {
"short_term": {"type": "short_term"},
"long_term": {"type": "long_term"},
"entities": {"type": "entity"},
"external": {"type": "external"},
}
if self.memory_type in memory_type_map:
params["metadata"] = memory_type_map[self.memory_type]
if self.memory_type == "short_term":
params["run_id"] = self.mem0_run_id
# Discard the filters for now since we create the filters
# automatically when the crew is created.
params["filters"] = self._create_filter_for_search()
params['threshold'] = score_threshold
if isinstance(self.memory, Memory):
del params["metadata"], params["output_format"]
del params["metadata"], params["version"], params["run_id"], params['output_format']
results = self.memory.search(**params)
return [r for r in results["results"] if r["score"] >= score_threshold]
def _get_user_id(self) -> str:
return self._get_config().get("user_id", "")
def _get_agent_name(self) -> str:
if not self.crew:
return ""
agents = self.crew.agents
agents = [self._sanitize_role(agent.role) for agent in agents]
agents = "_".join(agents)
return sanitize_collection_name(name=agents,max_collection_length=MAX_AGENT_ID_LENGTH_MEM0)
def _get_config(self) -> Dict[str, Any]:
return self.config or getattr(self, "memory_config", {}).get("config", {}) or {}
return [r for r in results["results"]]
def reset(self):
if self.memory:
self.memory.reset()

View File

@@ -7,8 +7,8 @@ import uuid
from typing import Any, Dict, List, Optional
from chromadb.api import ClientAPI
from crewai.memory.storage.base_rag_storage import BaseRAGStorage
from crewai.utilities import EmbeddingConfigurator
from crewai.rag.storage.base_rag_storage import BaseRAGStorage
from crewai.rag.embeddings.configurator import EmbeddingConfigurator
from crewai.utilities.chromadb import create_persistent_client
from crewai.utilities.constants import MAX_FILE_NAME_LENGTH
from crewai.utilities.paths import db_storage_path

View File

@@ -14,7 +14,8 @@ class UserMemory(Memory):
def __init__(self, crew=None):
warnings.warn(
"UserMemory is deprecated and will be removed in a future version. "
"UserMemory is deprecated and will be removed in version 0.156.0 "
"or on 2025-08-04, whichever comes first. "
"Please use ExternalMemory instead.",
DeprecationWarning,
stacklevel=2,

View File

@@ -1,8 +1,16 @@
import warnings
from typing import Any, Dict, Optional
class UserMemoryItem:
def __init__(self, data: Any, user: str, metadata: Optional[Dict[str, Any]] = None):
warnings.warn(
"UserMemoryItem is deprecated and will be removed in version 0.156.0 "
"or on 2025-08-04, whichever comes first. "
"Please use ExternalMemory instead.",
DeprecationWarning,
stacklevel=2,
)
self.data = data
self.user = user
self.metadata = metadata if metadata is not None else {}

View File

@@ -0,0 +1 @@
"""RAG (Retrieval-Augmented Generation) infrastructure for CrewAI."""

View File

@@ -0,0 +1 @@
"""Embedding components for RAG infrastructure."""

View File

@@ -0,0 +1 @@
"""Storage components for RAG infrastructure."""

View File

@@ -10,7 +10,6 @@ from .rpm_controller import RPMController
from .exceptions.context_window_exceeding_exception import (
LLMContextLengthExceededException,
)
from .embedding_configurator import EmbeddingConfigurator
__all__ = [
"Converter",
@@ -24,5 +23,4 @@ __all__ = [
"RPMController",
"YamlParser",
"LLMContextLengthExceededException",
"EmbeddingConfigurator",
]

View File

@@ -157,10 +157,6 @@ def get_llm_response(
from_agent=from_agent,
)
except Exception as e:
printer.print(
content=f"Error during LLM call: {e}",
color="red",
)
raise e
if not answer:
printer.print(
@@ -232,12 +228,17 @@ def handle_unknown_error(printer: Any, exception: Exception) -> None:
printer: Printer instance for output
exception: The exception that occurred
"""
error_message = str(exception)
if "litellm" in error_message:
return
printer.print(
content="An unknown error occurred. Please check the details below.",
color="red",
)
printer.print(
content=f"Error details: {exception}",
content=f"Error details: {error_message}",
color="red",
)

View File

@@ -1,6 +1,5 @@
from datetime import datetime
from datetime import datetime, timezone
from typing import Any, Dict, Optional
from pydantic import BaseModel, Field
from crewai.utilities.serialization import to_serializable
@@ -9,7 +8,7 @@ from crewai.utilities.serialization import to_serializable
class BaseEvent(BaseModel):
"""Base class for all events"""
timestamp: datetime = Field(default_factory=datetime.now)
timestamp: datetime = Field(default_factory=lambda: datetime.now(timezone.utc))
type: str
source_fingerprint: Optional[str] = None # UUID string of the source entity
source_type: Optional[str] = None # "agent", "task", "crew", "memory", "entity_memory", "short_term_memory", "long_term_memory", "external_memory"

View File

@@ -2010,7 +2010,6 @@ def test_crew_agent_executor_litellm_auth_error():
from litellm.exceptions import AuthenticationError
from crewai.agents.tools_handler import ToolsHandler
from crewai.utilities import Printer
# Create an agent and executor
agent = Agent(
@@ -2043,7 +2042,6 @@ def test_crew_agent_executor_litellm_auth_error():
# 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(
@@ -2057,13 +2055,6 @@ def test_crew_agent_executor_litellm_auth_error():
}
)
# 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()

View File

@@ -0,0 +1,209 @@
interactions:
- request:
body: '{"messages": [{"role": "user", "content": "What is the capital of France?"}],
"model": "o1-mini", "stop": ["stop"]}'
headers:
accept:
- application/json
accept-encoding:
- gzip, deflate, zstd
connection:
- keep-alive
content-length:
- '115'
content-type:
- application/json
host:
- api.openai.com
user-agent:
- OpenAI/Python 1.75.0
x-stainless-arch:
- arm64
x-stainless-async:
- 'false'
x-stainless-lang:
- python
x-stainless-os:
- MacOS
x-stainless-package-version:
- 1.75.0
x-stainless-raw-response:
- 'true'
x-stainless-read-timeout:
- '600.0'
x-stainless-retry-count:
- '0'
x-stainless-runtime:
- CPython
x-stainless-runtime-version:
- 3.11.12
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
body:
string: "{\n \"error\": {\n \"message\": \"Unsupported parameter: 'stop'
is not supported with this model.\",\n \"type\": \"invalid_request_error\",\n
\ \"param\": \"stop\",\n \"code\": \"unsupported_parameter\"\n }\n}"
headers:
CF-RAY:
- 961215744c94cb45-GIG
Connection:
- keep-alive
Content-Length:
- '196'
Content-Type:
- application/json
Date:
- Fri, 18 Jul 2025 12:46:46 GMT
Server:
- cloudflare
Set-Cookie:
- __cf_bm=KwJ1K47OHX4n2TZN8bMW37yKzKyK__S4HbTiCfyWjXM-1752842806-1.0.1.1-lweHFR7Kv2v7hT5I6xxYVz_7Ruu6aBdEgpJrSWrMxi_ficAeWC0oDeQ.0w2Lr1WRejIjqqcwSgdl6RixF2qEkjJZfS0pz_Vjjqexe44ayp4;
path=/; expires=Fri, 18-Jul-25 13:16:46 GMT; domain=.api.openai.com; HttpOnly;
Secure; SameSite=None
- _cfuvid=zv09c6bwcgNsYU80ah3wXzqeaIKyt_h61EAh_XRA87I-1752842806652-0.0.1.1-604800000;
path=/; domain=.api.openai.com; HttpOnly; Secure; SameSite=None
X-Content-Type-Options:
- nosniff
access-control-expose-headers:
- X-Request-ID
alt-svc:
- h3=":443"; ma=86400
cf-cache-status:
- DYNAMIC
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '20'
openai-project:
- proj_xitITlrFeen7zjNSzML82h9x
openai-version:
- '2020-10-01'
strict-transport-security:
- max-age=31536000; includeSubDomains; preload
x-envoy-upstream-service-time:
- '32'
x-ratelimit-limit-requests:
- '30000'
x-ratelimit-limit-tokens:
- '150000000'
x-ratelimit-remaining-requests:
- '29999'
x-ratelimit-remaining-tokens:
- '149999990'
x-ratelimit-reset-requests:
- 2ms
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_7be4715c3ee32aa406eacb68c7cc966e
status:
code: 400
message: Bad Request
- request:
body: '{"messages": [{"role": "user", "content": "What is the capital of France?"}],
"model": "o1-mini"}'
headers:
accept:
- application/json
accept-encoding:
- gzip, deflate, zstd
connection:
- keep-alive
content-length:
- '97'
content-type:
- application/json
cookie:
- __cf_bm=KwJ1K47OHX4n2TZN8bMW37yKzKyK__S4HbTiCfyWjXM-1752842806-1.0.1.1-lweHFR7Kv2v7hT5I6xxYVz_7Ruu6aBdEgpJrSWrMxi_ficAeWC0oDeQ.0w2Lr1WRejIjqqcwSgdl6RixF2qEkjJZfS0pz_Vjjqexe44ayp4;
_cfuvid=zv09c6bwcgNsYU80ah3wXzqeaIKyt_h61EAh_XRA87I-1752842806652-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
- OpenAI/Python 1.75.0
x-stainless-arch:
- arm64
x-stainless-async:
- 'false'
x-stainless-lang:
- python
x-stainless-os:
- MacOS
x-stainless-package-version:
- 1.75.0
x-stainless-raw-response:
- 'true'
x-stainless-read-timeout:
- '600.0'
x-stainless-retry-count:
- '0'
x-stainless-runtime:
- CPython
x-stainless-runtime-version:
- 3.11.12
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
body:
string: !!binary |
H4sIAAAAAAAAA3RSwU7jMBC95ytGPlYNakJhQ2/sgSsg7QUhFA32pJni2JHtwFao/76yC3XQwsWH
efOe35uZ9wJAsBIbELLHIIdRl78nGvaqOt/dPDxf71/fdg/9bXO3e5ETXt+LZWTY5x3J8Mk6k3YY
NQW25ghLRxgoqla/LupmXTeXqwQMVpGONFuVAxsu61W9LldXZVV/MHvLkrzYwGMBAPCe3ujRKPor
NpB0UmUg73FLYnNqAhDO6lgR6D37gCaIZQalNYFMsv2nJ5A4ckANtoMbh0YSsIfF4g4d+8XibM50
1E0eo3MzaT0D0BgbMCZPnp8+kMPJZceGfd86Qm9N/NkHO4qEHgqAp5R6+hJEjM4OY2iDfaEkW62P
ciLPOYPNJxhsQJ3rV83yG7VWUUDWfjY1IVH2pDIzjxgnxXYGFLNs/5v5TvuYm802q1yuf9TPgJQ0
BlLt6Eix/Jo4tzmKZ/hT22nIybHw5F5ZUhuYXFyEog4nfTwQ4fc+0NB2bLbkRsfpSuKui0PxDwAA
//8DAN7IUy8kAwAA
headers:
CF-RAY:
- 961216c3f9837e07-GRU
Connection:
- keep-alive
Content-Encoding:
- gzip
Content-Type:
- application/json
Date:
- Fri, 18 Jul 2025 12:47:41 GMT
Server:
- cloudflare
Transfer-Encoding:
- chunked
X-Content-Type-Options:
- nosniff
access-control-expose-headers:
- X-Request-ID
alt-svc:
- h3=":443"; ma=86400
cf-cache-status:
- DYNAMIC
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '1027'
openai-project:
- proj_xitITlrFeen7zjNSzML82h9x
openai-version:
- '2020-10-01'
strict-transport-security:
- max-age=31536000; includeSubDomains; preload
x-envoy-upstream-service-time:
- '1029'
x-ratelimit-limit-requests:
- '30000'
x-ratelimit-limit-tokens:
- '150000000'
x-ratelimit-remaining-requests:
- '29999'
x-ratelimit-remaining-tokens:
- '149999990'
x-ratelimit-reset-requests:
- 2ms
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_19a0763b09f0410b9d09598078a04cd6
status:
code: 200
message: OK
version: 1

View File

@@ -0,0 +1,206 @@
interactions:
- request:
body: '{"messages": [{"role": "user", "content": "What is the capital of France?"}],
"model": "o1-mini", "stop": ["stop"]}'
headers:
accept:
- application/json
accept-encoding:
- gzip, deflate, zstd
connection:
- keep-alive
content-length:
- '115'
content-type:
- application/json
cookie:
- __cf_bm=KwJ1K47OHX4n2TZN8bMW37yKzKyK__S4HbTiCfyWjXM-1752842806-1.0.1.1-lweHFR7Kv2v7hT5I6xxYVz_7Ruu6aBdEgpJrSWrMxi_ficAeWC0oDeQ.0w2Lr1WRejIjqqcwSgdl6RixF2qEkjJZfS0pz_Vjjqexe44ayp4;
_cfuvid=zv09c6bwcgNsYU80ah3wXzqeaIKyt_h61EAh_XRA87I-1752842806652-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
- OpenAI/Python 1.75.0
x-stainless-arch:
- arm64
x-stainless-async:
- 'false'
x-stainless-lang:
- python
x-stainless-os:
- MacOS
x-stainless-package-version:
- 1.75.0
x-stainless-raw-response:
- 'true'
x-stainless-read-timeout:
- '600.0'
x-stainless-retry-count:
- '0'
x-stainless-runtime:
- CPython
x-stainless-runtime-version:
- 3.11.12
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
body:
string: "{\n \"error\": {\n \"message\": \"Unsupported parameter: 'stop'
is not supported with this model.\",\n \"type\": \"invalid_request_error\",\n
\ \"param\": \"stop\",\n \"code\": \"unsupported_parameter\"\n }\n}"
headers:
CF-RAY:
- 961220323a627e05-GRU
Connection:
- keep-alive
Content-Length:
- '196'
Content-Type:
- application/json
Date:
- Fri, 18 Jul 2025 12:54:06 GMT
Server:
- cloudflare
X-Content-Type-Options:
- nosniff
access-control-expose-headers:
- X-Request-ID
alt-svc:
- h3=":443"; ma=86400
cf-cache-status:
- DYNAMIC
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '9'
openai-project:
- proj_xitITlrFeen7zjNSzML82h9x
openai-version:
- '2020-10-01'
strict-transport-security:
- max-age=31536000; includeSubDomains; preload
x-envoy-upstream-service-time:
- '11'
x-ratelimit-limit-requests:
- '30000'
x-ratelimit-limit-tokens:
- '150000000'
x-ratelimit-remaining-requests:
- '29999'
x-ratelimit-remaining-tokens:
- '149999990'
x-ratelimit-reset-requests:
- 2ms
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_e8d7880c5977029062d8487d215e5282
status:
code: 400
message: Bad Request
- request:
body: '{"messages": [{"role": "user", "content": "What is the capital of France?"}],
"model": "o1-mini"}'
headers:
accept:
- application/json
accept-encoding:
- gzip, deflate, zstd
connection:
- keep-alive
content-length:
- '97'
content-type:
- application/json
cookie:
- __cf_bm=KwJ1K47OHX4n2TZN8bMW37yKzKyK__S4HbTiCfyWjXM-1752842806-1.0.1.1-lweHFR7Kv2v7hT5I6xxYVz_7Ruu6aBdEgpJrSWrMxi_ficAeWC0oDeQ.0w2Lr1WRejIjqqcwSgdl6RixF2qEkjJZfS0pz_Vjjqexe44ayp4;
_cfuvid=zv09c6bwcgNsYU80ah3wXzqeaIKyt_h61EAh_XRA87I-1752842806652-0.0.1.1-604800000
host:
- api.openai.com
user-agent:
- OpenAI/Python 1.75.0
x-stainless-arch:
- arm64
x-stainless-async:
- 'false'
x-stainless-lang:
- python
x-stainless-os:
- MacOS
x-stainless-package-version:
- 1.75.0
x-stainless-raw-response:
- 'true'
x-stainless-read-timeout:
- '600.0'
x-stainless-retry-count:
- '0'
x-stainless-runtime:
- CPython
x-stainless-runtime-version:
- 3.11.12
method: POST
uri: https://api.openai.com/v1/chat/completions
response:
body:
string: !!binary |
H4sIAAAAAAAAA3SSQW/bMAyF7/4Vgo5BXCSeV6c5bkAPPTVbMaAYCoOT6JitLAkSPbQo8t8HKWns
Yu1FB3181HsUXwshJGm5FVL1wGrwpvw2In/fXY3Pcd/sftzf9ENvnurm569dc9/IZVK4P4+o+E11
odzgDTI5e8QqIDCmruvma7Wpv1T1ZQaD02iSzK3LgSyV1aqqy9VVua5Oyt6Rwii34nchhBCv+Uwe
rcZnuRWr5dvNgDHCHuX2XCSEDM6kGwkxUmSwLJcTVM4y2mz7rkehwBODEa4T1wGsQkFRLBa3ECgu
FhdzZcBujJCc29GYGQBrHUNKnj0/nMjh7LIjS7FvA0J0Nr0c2XmZ6aEQ4iGnHt8FkT64wXPL7glz
23V9bCenOc/h5kTZMZgZuKyWH/RrNTKQibO5SQWqRz1JpyHDqMnNQDFL97+dj3ofk5Pdz5xVm08f
mIBS6Bl16wNqUu9DT2UB0yZ+Vnaec7YsI4a/pLBlwpD+QmMHoznuiIwvkXFoO7J7DD5QXpT03cWh
+AcAAP//AwAo/zsSJwMAAA==
headers:
CF-RAY:
- 961220338bd47e05-GRU
Connection:
- keep-alive
Content-Encoding:
- gzip
Content-Type:
- application/json
Date:
- Fri, 18 Jul 2025 12:54:08 GMT
Server:
- cloudflare
Transfer-Encoding:
- chunked
X-Content-Type-Options:
- nosniff
access-control-expose-headers:
- X-Request-ID
alt-svc:
- h3=":443"; ma=86400
cf-cache-status:
- DYNAMIC
openai-organization:
- crewai-iuxna1
openai-processing-ms:
- '1280'
openai-project:
- proj_xitITlrFeen7zjNSzML82h9x
openai-version:
- '2020-10-01'
strict-transport-security:
- max-age=31536000; includeSubDomains; preload
x-envoy-upstream-service-time:
- '1286'
x-ratelimit-limit-requests:
- '30000'
x-ratelimit-limit-tokens:
- '150000000'
x-ratelimit-remaining-requests:
- '29999'
x-ratelimit-remaining-tokens:
- '149999990'
x-ratelimit-reset-requests:
- 2ms
x-ratelimit-reset-tokens:
- 0s
x-request-id:
- req_b7390d46fa4e14380d42162cb22045df
status:
code: 200
message: OK
version: 1

View File

@@ -755,3 +755,15 @@ def test_multiple_routers_from_same_trigger():
assert execution_order.index("anemia_analysis") > execution_order.index(
"anemia_router"
)
def test_flow_name():
class MyFlow(Flow):
name = "MyFlow"
@start()
def start(self):
return "Hello, world!"
flow = MyFlow()
assert flow.name == "MyFlow"

View File

@@ -1,3 +1,4 @@
import logging
import os
from time import sleep
from unittest.mock import MagicMock, patch
@@ -664,3 +665,49 @@ def test_handle_streaming_tool_calls_no_tools(mock_emit):
expected_completed_llm_call=1,
expected_final_chunk_result=response,
)
@pytest.mark.vcr(filter_headers=["authorization"])
def test_llm_call_when_stop_is_unsupported(caplog):
llm = LLM(model="o1-mini", stop=["stop"])
with caplog.at_level(logging.INFO):
result = llm.call("What is the capital of France?")
assert "Retrying LLM call without the unsupported 'stop'" in caplog.text
assert isinstance(result, str)
assert "Paris" in result
@pytest.mark.vcr(filter_headers=["authorization"])
def test_llm_call_when_stop_is_unsupported_when_additional_drop_params_is_provided(caplog):
llm = LLM(model="o1-mini", stop=["stop"], additional_drop_params=["another_param"])
with caplog.at_level(logging.INFO):
result = llm.call("What is the capital of France?")
assert "Retrying LLM call without the unsupported 'stop'" in caplog.text
assert isinstance(result, str)
assert "Paris" in result
@pytest.fixture
def ollama_llm():
return LLM(model="ollama/llama3.2:3b")
def test_ollama_appends_dummy_user_message_when_last_is_assistant(ollama_llm):
original_messages = [
{"role": "user", "content": "Hi there"},
{"role": "assistant", "content": "Hello!"},
]
formatted = ollama_llm._format_messages_for_provider(original_messages)
assert len(formatted) == len(original_messages) + 1
assert formatted[-1]["role"] == "user"
assert formatted[-1]["content"] == ""
def test_ollama_does_not_modify_when_last_is_user(ollama_llm):
original_messages = [
{"role": "user", "content": "Tell me a joke."},
]
formatted = ollama_llm._format_messages_for_provider(original_messages)
assert formatted == original_messages

View File

@@ -0,0 +1,142 @@
from crewai import Agent, Task, LLM
def test_langdb_basic_integration_example():
"""Test the basic LangDB integration example from the documentation."""
class MockLangDB:
@staticmethod
def init(**kwargs):
pass
MockLangDB.init()
llm = LLM(
model="gpt-4o",
temperature=0.7
)
agent = Agent(
role="Senior Research Analyst",
goal="Conduct comprehensive research on assigned topics",
backstory="You are an expert researcher with deep analytical skills.",
llm=llm,
verbose=True
)
assert agent.role == "Senior Research Analyst"
assert agent.goal == "Conduct comprehensive research on assigned topics"
assert agent.llm == llm
def test_langdb_metadata_configuration_example():
"""Test the metadata configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(metadata=None, **kwargs):
assert metadata is not None
assert "environment" in metadata
assert "crew_type" in metadata
MockLangDB.init(
metadata={
"environment": "production",
"crew_type": "research_workflow",
"user_id": "user_123"
}
)
def test_langdb_cost_tracking_example():
"""Test the cost tracking configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(cost_tracking=None, budget_alerts=None, **kwargs):
assert cost_tracking is True
assert budget_alerts is not None
assert "daily_limit" in budget_alerts
assert "alert_threshold" in budget_alerts
MockLangDB.init(
cost_tracking=True,
budget_alerts={
"daily_limit": 100.0,
"alert_threshold": 0.8
}
)
def test_langdb_security_configuration_example():
"""Test the security configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(security_config=None, **kwargs):
assert security_config is not None
assert "pii_detection" in security_config
assert "content_filtering" in security_config
assert "audit_logging" in security_config
MockLangDB.init(
security_config={
"pii_detection": True,
"content_filtering": True,
"audit_logging": True,
"data_retention_days": 90
}
)
def test_langdb_environment_specific_setup():
"""Test the multi-environment setup example from the documentation."""
environments = ["production", "staging", "development"]
for env in environments:
class MockLangDB:
@staticmethod
def init(project_id=None, sampling_rate=None, cost_tracking=None, **kwargs):
assert project_id is not None
assert sampling_rate is not None
assert cost_tracking is not None
if env == "production":
MockLangDB.init(
project_id="prod_project_id",
sampling_rate=1.0,
cost_tracking=True
)
elif env == "staging":
MockLangDB.init(
project_id="staging_project_id",
sampling_rate=0.5,
cost_tracking=False
)
else:
MockLangDB.init(
project_id="dev_project_id",
sampling_rate=0.1,
cost_tracking=False
)
def test_langdb_task_with_metadata():
"""Test task creation with metadata as shown in documentation."""
llm = LLM(model="gpt-4o")
agent = Agent(
role="Senior Research Analyst",
goal="Conduct research",
backstory="Expert researcher",
llm=llm
)
task = Task(
description="Research the latest AI trends",
expected_output="Comprehensive research report",
agent=agent
)
assert task.description == "Research the latest AI trends"
assert task.expected_output == "Comprehensive research report"
assert task.agent == agent

View File

@@ -0,0 +1,141 @@
"""Test for the LangDB documentation examples."""
from crewai import Agent, Task, LLM
def test_langdb_basic_integration_example():
"""Test the basic LangDB integration example from the documentation."""
class MockLangDB:
@staticmethod
def init(**kwargs):
pass
MockLangDB.init()
llm = LLM(
model="gpt-4o",
temperature=0.7
)
agent = Agent(
role="Senior Research Analyst",
goal="Conduct comprehensive research on assigned topics",
backstory="You are an expert researcher with deep analytical skills.",
llm=llm
)
assert agent.role == "Senior Research Analyst"
assert agent.goal == "Conduct comprehensive research on assigned topics"
assert agent.llm == llm
def test_langdb_metadata_configuration_example():
"""Test the metadata configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(metadata=None, **kwargs):
assert metadata is not None
assert "environment" in metadata
assert "crew_type" in metadata
MockLangDB.init(
metadata={
"environment": "production",
"crew_type": "research_workflow",
"user_id": "user_123"
}
)
def test_langdb_cost_tracking_example():
"""Test the cost tracking configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(cost_tracking=None, budget_alerts=None, **kwargs):
assert cost_tracking is True
assert budget_alerts is not None
assert "daily_limit" in budget_alerts
assert "alert_threshold" in budget_alerts
MockLangDB.init(
cost_tracking=True,
budget_alerts={
"daily_limit": 100.0,
"alert_threshold": 0.8
}
)
def test_langdb_security_configuration_example():
"""Test the security configuration example from the documentation."""
class MockLangDB:
@staticmethod
def init(security_config=None, **kwargs):
assert security_config is not None
assert "pii_detection" in security_config
assert "content_filtering" in security_config
assert "audit_logging" in security_config
MockLangDB.init(
security_config={
"pii_detection": True,
"content_filtering": True,
"audit_logging": True,
"data_retention_days": 90
}
)
def test_langdb_environment_specific_setup():
"""Test the multi-environment setup example from the documentation."""
environments = ["production", "staging", "development"]
for env in environments:
class MockLangDB:
@staticmethod
def init(project_id=None, sampling_rate=None, cost_tracking=None, **kwargs):
assert project_id is not None
assert sampling_rate is not None
assert cost_tracking is not None
if env == "production":
MockLangDB.init(
project_id="prod_project_id",
sampling_rate=1.0,
cost_tracking=True
)
elif env == "staging":
MockLangDB.init(
project_id="staging_project_id",
sampling_rate=0.5,
cost_tracking=False
)
else:
MockLangDB.init(
project_id="dev_project_id",
sampling_rate=0.1,
cost_tracking=False
)
def test_langdb_task_with_metadata():
"""Test task creation with metadata as shown in documentation."""
llm = LLM(model="gpt-4o")
agent = Agent(
role="Senior Research Analyst",
goal="Conduct research",
backstory="Expert researcher",
llm=llm
)
task = Task(
description="Research the latest AI trends",
expected_output="Comprehensive research report",
agent=agent
)
assert task.description == "Research the latest AI trends"
assert task.expected_output == "Comprehensive research report"
assert task.agent == agent

View File

@@ -1,14 +1,10 @@
import os
from unittest.mock import MagicMock, patch
import pytest
from mem0.client.main import MemoryClient
from mem0.memory.main import Memory
from crewai.agent import Agent
from crewai.crew import Crew
from crewai.memory.storage.mem0_storage import Mem0Storage
from crewai.task import Task
# Define the class (if not already defined)
@@ -59,10 +55,11 @@ def mem0_storage_with_mocked_config(mock_mem0_memory):
}
# Instantiate the class with memory_config
# Parameters like run_id, includes, and excludes doesn't matter in Memory OSS
crew = MockCrew(
memory_config={
"provider": "mem0",
"config": {"user_id": "test_user", "local_mem0_config": config},
"config": {"user_id": "test_user", "local_mem0_config": config, "run_id": "my_run_id", "includes": "include1","excludes": "exclude1", "infer" : True},
}
)
@@ -99,6 +96,10 @@ def mem0_storage_with_memory_client_using_config_from_crew(mock_mem0_memory_clie
"api_key": "ABCDEFGH",
"org_id": "my_org_id",
"project_id": "my_project_id",
"run_id": "my_run_id",
"includes": "include1",
"excludes": "exclude1",
"infer": True
},
}
)
@@ -154,11 +155,37 @@ def test_mem0_storage_with_explict_config(
assert (
mem0_storage_with_memory_client_using_explictly_config.config == expected_config
)
assert (
mem0_storage_with_memory_client_using_explictly_config.memory_config
== expected_config
def test_mem0_storage_updates_project_with_custom_categories(mock_mem0_memory_client):
mock_mem0_memory_client.update_project = MagicMock()
new_categories = [
{"lifestyle_management_concerns": "Tracks daily routines, habits, hobbies and interests including cooking, time management and work-life balance"},
]
crew = MockCrew(
memory_config={
"provider": "mem0",
"config": {
"user_id": "test_user",
"api_key": "ABCDEFGH",
"org_id": "my_org_id",
"project_id": "my_project_id",
"custom_categories": new_categories,
},
}
)
with patch.object(MemoryClient, "__new__", return_value=mock_mem0_memory_client):
_ = Mem0Storage(type="short_term", crew=crew)
mock_mem0_memory_client.update_project.assert_called_once_with(
custom_categories=new_categories
)
def test_save_method_with_memory_oss(mem0_storage_with_mocked_config):
"""Test save method for different memory types"""
@@ -172,9 +199,8 @@ def test_save_method_with_memory_oss(mem0_storage_with_mocked_config):
mem0_storage.save(test_value, test_metadata)
mem0_storage.memory.add.assert_called_once_with(
test_value,
agent_id="Test_Agent",
infer=False,
[{'role': 'assistant' , 'content': test_value}],
infer=True,
metadata={"type": "short_term", "key": "value"},
)
@@ -191,11 +217,14 @@ def test_save_method_with_memory_client(mem0_storage_with_memory_client_using_co
mem0_storage.save(test_value, test_metadata)
mem0_storage.memory.add.assert_called_once_with(
test_value,
agent_id="Test_Agent",
infer=False,
[{'role': 'assistant' , 'content': test_value}],
infer=True,
metadata={"type": "short_term", "key": "value"},
output_format="v1.1"
version="v2",
run_id="my_run_id",
includes="include1",
excludes="exclude1",
output_format='v1.1'
)
@@ -210,11 +239,12 @@ def test_search_method_with_memory_oss(mem0_storage_with_mocked_config):
mem0_storage.memory.search.assert_called_once_with(
query="test query",
limit=5,
agent_id="Test_Agent",
user_id="test_user"
user_id="test_user",
filters={'AND': [{'run_id': 'my_run_id'}]},
threshold=0.5
)
assert len(results) == 1
assert len(results) == 2
assert results[0]["content"] == "Result 1"
@@ -229,11 +259,31 @@ def test_search_method_with_memory_client(mem0_storage_with_memory_client_using_
mem0_storage.memory.search.assert_called_once_with(
query="test query",
limit=5,
agent_id="Test_Agent",
metadata={"type": "short_term"},
user_id="test_user",
output_format='v1.1'
version='v2',
run_id="my_run_id",
output_format='v1.1',
filters={'AND': [{'run_id': 'my_run_id'}]},
threshold=0.5
)
assert len(results) == 1
assert len(results) == 2
assert results[0]["content"] == "Result 1"
def test_mem0_storage_default_infer_value(mock_mem0_memory_client):
"""Test that Mem0Storage sets infer=True by default for short_term memory."""
with patch.object(MemoryClient, "__new__", return_value=mock_mem0_memory_client):
crew = MockCrew(
memory_config={
"provider": "mem0",
"config": {
"user_id": "test_user",
"api_key": "ABCDEFGH"
},
}
)
mem0_storage = Mem0Storage(type="short_term", crew=crew)
assert mem0_storage.infer is True

View File

@@ -64,7 +64,8 @@ def base_agent():
llm="gpt-4o-mini",
goal="Just say hi",
backstory="You are a helpful assistant that just says hi",
)
)
@pytest.fixture(scope="module")
def base_task(base_agent):
@@ -74,6 +75,7 @@ def base_task(base_agent):
agent=base_agent,
)
event_listener = EventListener()
@@ -448,6 +450,27 @@ def test_flow_emits_start_event():
assert received_events[0].type == "flow_started"
def test_flow_name_emitted_to_event_bus():
received_events = []
class MyFlowClass(Flow):
name = "PRODUCTION_FLOW"
@start()
def start(self):
return "Hello, world!"
@crewai_event_bus.on(FlowStartedEvent)
def handle_flow_start(source, event):
received_events.append(event)
flow = MyFlowClass()
flow.kickoff()
assert len(received_events) == 1
assert received_events[0].flow_name == "PRODUCTION_FLOW"
def test_flow_emits_finish_event():
received_events = []
@@ -756,6 +779,7 @@ def test_streaming_empty_response_handling():
received_chunks = []
with crewai_event_bus.scoped_handlers():
@crewai_event_bus.on(LLMStreamChunkEvent)
def handle_stream_chunk(source, event):
received_chunks.append(event.chunk)
@@ -793,6 +817,7 @@ def test_streaming_empty_response_handling():
# Restore the original method
llm.call = original_call
@pytest.mark.vcr(filter_headers=["authorization"])
def test_stream_llm_emits_event_with_task_and_agent_info():
completed_event = []
@@ -801,6 +826,7 @@ def test_stream_llm_emits_event_with_task_and_agent_info():
stream_event = []
with crewai_event_bus.scoped_handlers():
@crewai_event_bus.on(LLMCallFailedEvent)
def handle_llm_failed(source, event):
failed_event.append(event)
@@ -827,7 +853,7 @@ def test_stream_llm_emits_event_with_task_and_agent_info():
description="Just say hi",
expected_output="hi",
llm=LLM(model="gpt-4o-mini", stream=True),
agent=agent
agent=agent,
)
crew = Crew(agents=[agent], tasks=[task])
@@ -855,6 +881,7 @@ def test_stream_llm_emits_event_with_task_and_agent_info():
assert set(all_task_id) == {task.id}
assert set(all_task_name) == {task.name}
@pytest.mark.vcr(filter_headers=["authorization"])
def test_llm_emits_event_with_task_and_agent_info(base_agent, base_task):
completed_event = []
@@ -863,6 +890,7 @@ def test_llm_emits_event_with_task_and_agent_info(base_agent, base_task):
stream_event = []
with crewai_event_bus.scoped_handlers():
@crewai_event_bus.on(LLMCallFailedEvent)
def handle_llm_failed(source, event):
failed_event.append(event)
@@ -904,6 +932,7 @@ def test_llm_emits_event_with_task_and_agent_info(base_agent, base_task):
assert set(all_task_id) == {base_task.id}
assert set(all_task_name) == {base_task.name}
@pytest.mark.vcr(filter_headers=["authorization"])
def test_llm_emits_event_with_lite_agent():
completed_event = []
@@ -912,6 +941,7 @@ def test_llm_emits_event_with_lite_agent():
stream_event = []
with crewai_event_bus.scoped_handlers():
@crewai_event_bus.on(LLMCallFailedEvent)
def handle_llm_failed(source, event):
failed_event.append(event)
@@ -936,7 +966,6 @@ def test_llm_emits_event_with_lite_agent():
)
agent.kickoff(messages=[{"role": "user", "content": "say hi!"}])
assert len(completed_event) == 2
assert len(failed_event) == 0
assert len(started_event) == 2

6110
uv.lock generated

File diff suppressed because it is too large Load Diff