frameworks

LongMemory agent framework examples

Runnable examples connect popular Python agent frameworks to the installed LongMemory CLI through each framework's native MCP client.

Framework	Native integration
CrewAI	Agent.mcps with MCPServerStdio
AutoGen	McpWorkbench with StdioServerParams
LangGraph/LangChain	MultiServerMCPClient tools passed to create_agent
OpenAI Agents SDK	MCPServerStdio attached through Agent.mcp_servers
PydanticAI	MCPToolset backed by FastMCP StdioTransport

Run an example

Install and initialize the LongMemory CLI first. Then create an isolated Python environment inside the chosen example directory:

cd integrations/frameworks/crewai
py -3.11 -m venv .venv
.\.venv\Scripts\python -m pip install -r requirements.txt
.\.venv\Scripts\python main.py

Use the same steps for autogen, langgraph, openai-agents, or pydantic-ai.

Each example accepts:

• LONGMEMORY_COMMAND: override the LongMemory executable; defaults to longmemory.cmd on Windows and longmemory elsewhere.
• LONGMEMORY_MODEL: override the framework-specific model identifier.
• LONGMEMORY_TASK: replace the default project-context request.

The model provider's own credentials are still required by the selected framework. No credential belongs in these files.

Deployment boundary

The examples use local stdio so the framework owns the LongMemory subprocess lifecycle. For a hosted or multi-user deployment, use the framework's Streamable HTTP MCP class with longmemory serve --mcp-http, bearer authentication, and a server-bound tenant, user, project, agent, and framework identity. Never derive those identities from model-generated tool arguments.