fixes from discussion

docs/concepts/knowledge.mdx

@@ -51,7 +51,7 @@ crew = Crew(
     tasks=[task],
     verbose=True,
     process=Process.sequential,
-    knowledge={"sources": [string_source], "metadata": {"preference": "personal"}}, # Enable knowledge by adding the sources here. You can also add more sources to the sources list.
+    knowledge_sources=[string_source], # Enable knowledge by adding the sources here. You can also add more sources to the sources list.
 )
 
 result = crew.kickoff(inputs={"question": "What city does John live in and how old is he?"})
@@ -63,18 +63,29 @@ Sometimes you may want to append knowledge sources to an individual agent. You c
 ```python
 agent = Agent(
     ...
-    knowledge={
+    knowledge_sources=[
-        "sources": [
+        StringKnowledgeSource(
-            StringKnowledgeSource(
+            content="Users name is John. He is 30 years old and lives in San Francisco.",
-                content="Users name is John. He is 30 years old and lives in San Francisco.",
+            metadata={"preference": "personal"},
-                metadata={"preference": "personal"},
+        )
-            )
+    ],
-        ],
-        "metadata": {"preference": "personal"},
-    },
 )
 ```
 
+## Agent Level Knowledge Sources
+
+You can also append knowledge sources to an individual agent by setting the `knowledge_sources` parameter in the `Agent` class.
+
+```python
+string_source = StringKnowledgeSource(
+    content="Users name is John. He is 30 years old and lives in San Francisco.",
+    metadata={"preference": "personal"},
+)
+agent = Agent(
+    ...
+    knowledge_sources=[string_source],
+)
+```
 
 ## Embedder Configuration
 
@@ -88,10 +99,7 @@ string_source = StringKnowledgeSource(
 )
 crew = Crew(
     ...
-    knowledge={
+    knowledge_sources=[string_source],
-        "sources": [string_source],
+    embedder_config={"provider": "ollama", "config": {"model": "nomic-embed-text:latest"}},
-        "metadata": {"preference": "personal"},
-        "embedder_config": {"provider": "openai", "config": {"model": "text-embedding-3-small"}},
-    },
 )
 ```

src/crewai/agent.py

@@ -21,6 +21,7 @@ from crewai.utilities.constants import TRAINED_AGENTS_DATA_FILE, TRAINING_DATA_F
 from crewai.utilities.converter import generate_model_description
 from crewai.utilities.token_counter_callback import TokenCalcHandler
 from crewai.utilities.training_handler import CrewTrainingHandler
+from crewai.knowledge.utils.knowledge_utils import extract_knowledge_context
 
 
 def mock_agent_ops_provider():
@@ -124,14 +125,17 @@ class Agent(BaseAgent):
         default="safe",
         description="Mode for code execution: 'safe' (using Docker) or 'unsafe' (direct execution).",
     )
-    knowledge: Optional[Union[List[BaseKnowledgeSource], Knowledge]] = Field(
-        default=None,
-        description="Knowledge for the agent. Add knowledge sources to the knowledge object.",
-    )
     embedder_config: Optional[Dict[str, Any]] = Field(
         default=None,
         description="Embedder configuration for the agent.",
     )
+    knowledge_sources: Optional[List[BaseKnowledgeSource]] = Field(
+        default=None,
+        description="Knowledge sources for the agent. Add knowledge sources to the knowledge object.",
+    )
+    _knowledge: Optional[Knowledge] = PrivateAttr(
+        default=None,
+    )
 
     @model_validator(mode="after")
     def post_init_setup(self):
@@ -245,14 +249,13 @@ class Agent(BaseAgent):
 
     def _set_knowledge(self):
         try:
-            if self.knowledge:
+            if self.knowledge_sources:
                 knowledge_agent_name = f"{self.role.replace(' ', '_')}"
-                print("knowledge_agent_name", knowledge_agent_name)
+                if isinstance(self.knowledge_sources, list) and all(
-                if isinstance(self.knowledge, list) and all(
+                    isinstance(k, BaseKnowledgeSource) for k in self.knowledge_sources
-                    isinstance(k, BaseKnowledgeSource) for k in self.knowledge
                 ):
-                    self.knowledge = Knowledge(
+                    self._knowledge = Knowledge(
-                        sources=self.knowledge,
+                        sources=self.knowledge_sources,
                         embedder_config=self.embedder_config,
                         collection_name=knowledge_agent_name,
                     )
@@ -313,22 +316,21 @@ class Agent(BaseAgent):
             if memory.strip() != "":
                 task_prompt += self.i18n.slice("memory").format(memory=memory)
 
-        if self.knowledge and isinstance(self.knowledge, Knowledge):
+        if self._knowledge:
-            agent_knowledge_snippets = self.knowledge.query([task.prompt()])
+            agent_knowledge_snippets = self._knowledge.query([task.prompt()])
-            agent_knowledge_context = self.knowledge.extract_knowledge_context(
+            if agent_knowledge_snippets:
-                agent_knowledge_snippets
+                agent_knowledge_context = extract_knowledge_context(
-            )
+                    agent_knowledge_snippets
-            if agent_knowledge_context:
+                )
-                task_prompt += agent_knowledge_context
+                if agent_knowledge_context:
+                    task_prompt += agent_knowledge_context
 
-        if self.crew and self.crew.knowledge:
+        if self.crew:
-            knowledge_snippets = self.crew.knowledge.query([task.prompt()])
+            knowledge_snippets = self.crew.query_knowledge([task.prompt()])
-
+            if knowledge_snippets:
-            crew_knowledge_context = self.crew.knowledge.extract_knowledge_context(
+                crew_knowledge_context = extract_knowledge_context(knowledge_snippets)
-                knowledge_snippets
+                if crew_knowledge_context:
-            )
+                    task_prompt += crew_knowledge_context
-            if crew_knowledge_context:
-                task_prompt += crew_knowledge_context
 
         tools = tools or self.tools or []
         self.create_agent_executor(tools=tools, task=task)

src/crewai/crew.py

@@ -203,9 +203,12 @@ class Crew(BaseModel):
         default=[],
         description="List of execution logs for tasks",
     )
-    knowledge: Optional[Union[List[BaseKnowledgeSource], Knowledge]] = Field(
+    knowledge_sources: Optional[List[BaseKnowledgeSource]] = Field(
+        default=None,
+        description="Knowledge sources for the crew. Add knowledge sources to the knowledge object.",
+    )
+    _knowledge: Optional[Knowledge] = PrivateAttr(
         default=None,
-        description="Knowledge for the crew. Add knowledge sources to the knowledge object.",
     )
 
     @field_validator("id", mode="before")
@@ -284,13 +287,13 @@ class Crew(BaseModel):
     @model_validator(mode="after")
     def create_crew_knowledge(self) -> "Crew":
         """Create the knowledge for the crew."""
-        if self.knowledge:
+        if self.knowledge_sources:
             try:
-                if isinstance(self.knowledge, list) and all(
+                if isinstance(self.knowledge_sources, list) and all(
-                    isinstance(k, BaseKnowledgeSource) for k in self.knowledge
+                    isinstance(k, BaseKnowledgeSource) for k in self.knowledge_sources
                 ):
-                    self.knowledge = Knowledge(
+                    self._knowledge = Knowledge(
-                        sources=self.knowledge,
+                        sources=self.knowledge_sources,
                         embedder_config=self.embedder,
                         collection_name="crew",
                     )
@@ -954,6 +957,11 @@ class Crew(BaseModel):
         result = self._execute_tasks(self.tasks, start_index, True)
         return result
 
+    def query_knowledge(self, query: List[str]) -> Union[List[Dict[str, Any]], None]:
+        if self._knowledge:
+            return self._knowledge.query(query)
+        return None
+
     def copy(self):
         """Create a deep copy of the Crew."""
 

src/crewai/knowledge/knowledge.py

@@ -62,18 +62,6 @@ class Knowledge(BaseModel):
         )
         return results
 
-    def extract_knowledge_context(
-        self, knowledge_snippets: List[Dict[str, Any]]
-    ) -> str:
-        """Extract knowledge from the task prompt."""
-        valid_snippets = [
-            result["context"]
-            for result in knowledge_snippets
-            if result and result.get("context")
-        ]
-        snippet = "\n".join(valid_snippets)
-        return f"Additional Information: {snippet}" if valid_snippets else ""
-
     def _add_sources(self):
         for source in self.sources:
             source.storage = self.storage

src/crewai/knowledge/storage/knowledge_storage.py

@@ -3,12 +3,16 @@ import io
 import logging
 import chromadb
 import os
+
+import chromadb.errors
 from crewai.utilities.paths import db_storage_path
-from typing import Optional, List, Dict, Any
+from typing import Optional, List, Dict, Any, Union
 from crewai.utilities import EmbeddingConfigurator
 from crewai.knowledge.storage.base_knowledge_storage import BaseKnowledgeStorage
 import hashlib
 from chromadb.config import Settings
+from chromadb.api import ClientAPI
+from crewai.utilities.logger import Logger
 
 
 @contextlib.contextmanager
@@ -36,15 +40,15 @@ class KnowledgeStorage(BaseKnowledgeStorage):
 
     collection: Optional[chromadb.Collection] = None
     collection_name: Optional[str] = "knowledge"
-    app: Optional[chromadb.PersistentClient] = None
+    app: Optional[ClientAPI] = None
 
     def __init__(
         self,
         embedder_config: Optional[Dict[str, Any]] = None,
         collection_name: Optional[str] = None,
     ):
-        self.embedder_config = embedder_config
         self.collection_name = collection_name
+        self._set_embedder_config(embedder_config)
 
     def search(
         self,
@@ -91,7 +95,7 @@ class KnowledgeStorage(BaseKnowledgeStorage):
             )
             if self.app:
                 self.collection = self.app.get_or_create_collection(
-                    name=collection_name
+                    name=collection_name, embedding_function=self.embedder_config
                 )
             else:
                 raise Exception("Vector Database Client not initialized")
@@ -110,18 +114,39 @@ class KnowledgeStorage(BaseKnowledgeStorage):
             self.app.reset()
 
     def save(
-        self, documents: List[str], metadata: Dict[str, Any] | List[Dict[str, Any]]
+        self,
+        documents: List[str],
+        metadata: Union[Dict[str, Any], List[Dict[str, Any]]],
     ):
         if self.collection:
-            metadatas = [metadata] if isinstance(metadata, dict) else metadata
+            try:
+                metadatas = [metadata] if isinstance(metadata, dict) else metadata
 
-            ids = [hashlib.sha256(doc.encode("utf-8")).hexdigest() for doc in documents]
+                ids = [
+                    hashlib.sha256(doc.encode("utf-8")).hexdigest() for doc in documents
+                ]
 
-            self.collection.upsert(
+                self.collection.upsert(
-                documents=documents,
+                    documents=documents,
-                metadatas=metadatas,
+                    metadatas=metadatas,
-                ids=ids,
+                    ids=ids,
-            )
+                )
+            except chromadb.errors.InvalidDimensionException as e:
+                Logger(verbose=True).log(
+                    "error",
+                    "Embedding dimension mismatch. This usually happens when mixing different embedding models. Try resetting the collection using `crewai reset-memories -a`",
+                    "red",
+                )
+                raise ValueError(
+                    "Embedding dimension mismatch. Make sure you're using the same embedding model "
+                    "across all operations with this collection."
+                    "Try resetting the collection using `crewai reset-memories -a`"
+                ) from e
+            except Exception as e:
+                Logger(verbose=True).log(
+                    "error", f"Failed to upsert documents: {e}", "red"
+                )
+                raise
         else:
             raise Exception("Collection not initialized")
 

src/crewai/knowledge/utils/knowledge_utils.py

@@ -0,0 +1,12 @@
+from typing import Any, Dict, List
+
+
+def extract_knowledge_context(knowledge_snippets: List[Dict[str, Any]]) -> str:
+    """Extract knowledge from the task prompt."""
+    valid_snippets = [
+        result["context"]
+        for result in knowledge_snippets
+        if result and result.get("context")
+    ]
+    snippet = "\n".join(valid_snippets)
+    return f"Additional Information: {snippet}" if valid_snippets else ""

tests/agent_test.py

@@ -3,7 +3,6 @@
 import os
 from unittest import mock
 from unittest.mock import patch
-
 import pytest
 
 from crewai import Agent, Crew, Task
@@ -11,7 +10,6 @@ from crewai.knowledge.source.base_knowledge_source import BaseKnowledgeSource
 from crewai.agents.cache import CacheHandler
 from crewai.agents.crew_agent_executor import CrewAgentExecutor
 from crewai.agents.parser import AgentAction, CrewAgentParser, OutputParserException
-from crewai.knowledge.knowledge import Knowledge
 from crewai.llm import LLM
 from crewai.tools import tool
 from crewai.tools.tool_calling import InstructorToolCalling
@@ -1627,10 +1625,9 @@ def test_agent_with_knowledge_sources_context():
         goal="Provide information based on knowledge sources",
         backstory="You have access to specific knowledge sources.",
         llm=LLM(model="gpt-4o-mini"),
-        knowledge=[string_source],
+        knowledge_sources=[string_source],
     )
-    # Test that agent is properly initialized with knowledge sources
+    assert isinstance(agent.knowledge_sources, list)
-    assert isinstance(agent.knowledge, Knowledge)
+    assert len(agent.knowledge_sources) == 1
-    assert len(agent.knowledge.sources) == 1
+    assert isinstance(agent.knowledge_sources[0], BaseKnowledgeSource)
-    assert isinstance(agent.knowledge.sources[0], BaseKnowledgeSource)
+    assert agent.knowledge_sources[0].metadata == {"preference": "personal"}
-    assert agent.knowledge.sources[0].metadata == {"preference": "personal"}