Agents & Memory

Integrate Endee with CrewAI to give your agents persistent vector-based memory for contextual reasoning and knowledge retrieval.

Memory Types

CrewAI supports different memory types that can use Endee as a backend:

Memory Type	Description
ShortTermMemory	Recent context and conversation history
EntityMemory	Information about entities (people, places, concepts)

Creating CrewAI Memory Objects

Use your Endee vector store as the storage backend for CrewAI memory:


from crewai.memory import ShortTermMemory, EntityMemory
 
# Create CrewAI memory objects with Endee backend
short_memory = ShortTermMemory(storage=memory_store)
entity_memory = EntityMemory(storage=memory_store)

Defining an Agent with Memory

Create an agent that uses Endee-backed memory:


from crewai import Agent, LLM
 
# Define LLM (use any valid model)
llm = LLM(model="gemini-2.5-flash-lite", api_key="<GOOGLE_API_KEY>")
 
# Define an agent with memory
agent = Agent(
    role="Programming Language Expert",
    goal="Answer questions using stored programming language knowledge.",
    backstory="Consult documents and provide concise answers.",
    llm=llm,
    memory=short_memory,
    verbose=True
)

Defining Tasks

Create tasks for your agents:


from crewai import Task
 
# Define a task
task = Task(
    description="Answer questions about programming languages.",
    agent=agent,
    expected_output="Concise and accurate answers."
)

Running a Crew

Bring it all together with a Crew:


from crewai import Crew, Process
 
# Run Crew with memory
crew = Crew(
    agents=[agent],
    tasks=[task],
    process=Process.sequential,
    memory=True,
    short_term_memory=short_memory,
    entity_memory=entity_memory,
    verbose=True
)
 
result = crew.kickoff()
print(result)

Complete Example

Here’s a full working example:


from endee_crewai import EndeeVectorStore
from crewai.memory import ShortTermMemory, EntityMemory
from crewai import Agent, Crew, Task, Process, LLM
import time
 
# Initialize Endee vector store
embedder_config = {
    "provider": "cohere",
    "config": {"model_name": "small", "api_key": "<COHERE_API_KEY>"}
}
 
memory_store = EndeeVectorStore(
    type="programming_knowledge",
    api_token="<ENDEE_API_TOKEN>",
    embedder_config=embedder_config,
    space_type="cosine",
    crew=None,
)
 
# Reset and populate with knowledge
memory_store.reset()
time.sleep(2)
 
documents = [
    ("Python is dynamically typed.", {"creator": "Guido van Rossum", "typing": "dynamic"}),
    ("Go is statically typed with great concurrency.", {"creator": "Robert Griesemer", "typing": "static"}),
    ("JavaScript is the language of the web.", {"creator": "Brendan Eich", "typing": "dynamic"}),
]
 
for text, meta in documents:
    memory_store.save(text, meta)
 
time.sleep(1)
 
# Create CrewAI memory objects
short_memory = ShortTermMemory(storage=memory_store)
entity_memory = EntityMemory(storage=memory_store)
 
# Define LLM
llm = LLM(model="gemini-2.5-flash-lite", api_key="<GOOGLE_API_KEY>")
 
# Define agent
agent = Agent(
    role="Programming Language Expert",
    goal="Answer questions using stored programming language knowledge.",
    backstory="Consult documents and provide concise answers.",
    llm=llm,
    memory=short_memory,
    verbose=True
)
 
# Define task
task = Task(
    description="What are the differences between Python and Go?",
    agent=agent,
    expected_output="A concise comparison of Python and Go."
)
 
# Run Crew
crew = Crew(
    agents=[agent],
    tasks=[task],
    process=Process.sequential,
    memory=True,
    short_term_memory=short_memory,
    entity_memory=entity_memory,
    verbose=True
)
 
result = crew.kickoff()
print(result)

Helper Functions

Create helper functions to validate your memory store:


def check_language_vectors(memory_store):
    """Check what's stored in the vector memory."""
    queries = ["Python language features", "Go concurrency"]
    for q in queries:
        results = memory_store.search(query=q, limit=3)
        for r in results:
            print(r["context"])
 
# Test the memory store
check_language_vectors(memory_store)

Next Steps

Explore configuration options