Description
Description
The test code:
from crewai import Agent, Task, Crew, Process, LLM
from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
# Create a knowledge source
content = "Users name is John. He is 30 years old and lives in San Francisco."
string_source = StringKnowledgeSource(
content=content,
)
# Create an LLM with a temperature of 0 to ensure deterministic outputs
llm = LLM(model="gpt-4o", temperature=0)
# Create an agent with the knowledge store
agent = Agent(
role="About User",
goal="You know everything about the user.",
backstory="""You are a master at understanding people and their preferences.""",
verbose=True,
allow_delegation=False,
llm=llm,
knowledge_sources=[string_source],
)
task = Task(
description="Answer the following questions about the user: {question}",
expected_output="An answer to the question.",
agent=agent,
)
crew = Crew(
agents=[agent],
tasks=[task],
verbose=True,
process=Process.sequential,
knowledge_sources=[string_source],
)
result = crew.kickoff(inputs={"question": "What city does John live in and how old is he?"})
Firstly, i use knowledge_sources on both Agent and Crew, but then i check my database, only knowledge of Crew is added to chromadb, Agent's knowledge isn't. So i check set_knowledge function in agent.py and added: self.knowledge.add_sources() , it works.
Secondly, if i only define knowledge_sources on Crew, the Agent actually doesnt get access to knowledge source. Continue checking, i realize that in this logic, knowledge of Crew only add to task_prompt when knowledge of Agent exist. I thought if i define knowledge_sources on Crew level, all Agents will get access to the knowledge as well?
Are these bugs or what am i missing here?
Steps to Reproduce
Run the test code
Expected behavior
Knowledges of Agent should be added to database and knowledge_sources should work on Crew level
Screenshots/Code snippets
Database's collection when run test code (Collection of Agent is empty):
Collection: knowledge_crew
{'ids': ['225f7a909030abdbcdd1f8d0657e0e5bcd64a725936f50246ec4ac28fcf4ffe9'], 'embeddings': None, 'documents': ['Users name is John. He is 30 years old and lives in San Francisco.'], 'uris': None, 'data': None, 'metadatas': [None], 'included': [<IncludeEnum.documents: 'documents'>, <IncludeEnum.metadatas: 'metadatas'>]}
Collection: knowledge_About_User
{'ids': [], 'embeddings': None, 'documents': [], 'uris': None, 'data': None, 'metadatas': [], 'included': [<IncludeEnum.documents: 'documents'>, <IncludeEnum.metadatas: 'metadatas'>]}
Operating System
Ubuntu 20.04
Python Version
3.12
crewAI Version
0.119.0
crewAI Tools Version
0.119.0
Virtual Environment
Venv
Evidence
None
Possible Solution
None
Additional context
None