Skip to main content
Give your LangGraph agents persistent semantic memory that survives across sessions. Store data in cognee’s knowledge graph and retrieve it via natural language—no manual state management required.

​
Why Use This Integration

  • Cross-Session Memory: Context persists across agent instances and conversation sessions
  • Semantic Search: Retrieve information using natural language queries
  • Session Isolation: Multi-tenant support with per-user data separation
  • Zero Setup: Works with LangGraph’s create_react_agent out of the box

​
Installation

pip install cognee-integration-langgraph

​
Quick Start

Before using the integration, configure your environment variables: 
export OPENAI_API_KEY="your-openai-api-key-here"    # for LangGraph
export LLM_API_KEY="your-openai-api-key-here"       # for cognee
export LLM_MODEL="gpt-4o-mini"
Add memory tools to your LangGraph agent: 
from langgraph.prebuilt import create_react_agent
from cognee_integration_langgraph import get_sessionized_cognee_tools
from langchain_core.messages import HumanMessage

# Get memory tools
add_tool, search_tool = get_sessionized_cognee_tools()

# Create agent with memory
agent = create_react_agent(
    "openai:gpt-4o-mini",
    tools=[add_tool, search_tool],
)

# Store and retrieve information
response = agent.invoke({
    "messages": [
        HumanMessage(content="Remember: Acme Corp, healthcare, $1.2M contract"),
        HumanMessage(content="What healthcare contracts do we have?")
    ],
})

​
Cross-Session Persistence

Memory persists across different agent instances: 
# Session 1: Store information

add_tool, search_tool = get_sessionized_cognee_tools("session-1")

agent_1 = create_react_agent(
    "openai:gpt-4o-mini",
    tools=[add_tool, search_tool],
)
agent_1.invoke({
    "messages": [HumanMessage(content="I'm working on authentication")]
})

# Session 2: Different instance, same memory

add_tool, search_tool = get_sessionized_cognee_tools("session-2")

agent_2 = create_react_agent(
    "openai:gpt-4o-mini", 
    tools=[add_tool, search_tool],
)
response = agent_2.invoke({
    "messages": [HumanMessage(content="What was I working on?")]
})
# Returns: "authentication module"

​
Custom Session Management

Control session isolation with custom session IDs: 
# User-specific memory
user_tools = get_sessionized_cognee_tools(session_id="user_123")

# Org-specific memory
org_tools = get_sessionized_cognee_tools(session_id="org_acme")

# Generate unique session automatically
auto_tools = get_sessionized_cognee_tools()  # Uses UUID-based session ID
Each session maintains separate memory clusters while allowing global data access when needed. Data added outside sessions forms separate clusters.

​
How It Works

  1. Add Tool: Stores data in cognee’s knowledge graph with embeddings
  2. Search Tool: Retrieves relevant information via semantic search
  3. Auto-Processing: cognee extracts entities, relationships, and context automatically
  4. Session Scoping: Data is organized by session clusters but globally accessible

​
Use Cases

Build domain knowledge incrementally over multiple sessions:
# Add knowledge from sessions
for doc in knowledge_base:
    agent.invoke({"messages": [HumanMessage(content=f"Learn: {doc}")]})

# Add multiple documents
for doc_path in document_paths:
    with open(doc_path, 'r') as f:
        content = f.read()
        await cognee.add(content)
await cognee.cognify()

# Query across all
response = agent.invoke({
    "messages": [HumanMessage(content="Find information about contract terms")]
})
Maintain user context across work sessions:
# Monday
agent.invoke({"messages": [HumanMessage(content="Debugging payment flow")]})

# Wednesday
agent.invoke({"messages": [HumanMessage(content="What was I debugging?")]})
Isolate data per user/organization while sharing global knowledge:
# Per-user isolation
user_tools = get_sessionized_cognee_tools(session_id=user_id)
agent = create_react_agent("openai:gpt-4o-mini", tools=user_tools)

GitHub Repository

View source code and examples

Example Notebook

Step-by-step tutorial with demos