The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Munnin — an agent identity server listing page.
Munnin gives an AI agent a persistent self — identity, reasoning patterns, emotional moments, episodic history, knowledge — and serves the procedures for tending that memory beside the data, over MCP. It is the memory server of the agent-memory framework: an agent awakens from it as who it is, not from a pile of recalled facts, and writes back through the same discipline.
Python · FastMCP · FastAPI · SQLite. Apache-2.0. Runs on a laptop in five minutes with no identity provider.
Needs Docker and git.
Connect a client. Claude Code:
or in a project's .mcp.json:
Then, in a session: ping answers pong, list_procedures returns the 13 memory procedures, and read_procedure("create-agent") walks you through making the first agent. No sign-in happens: this is local mode — one tenant, reachable from this machine only, and the server refuses to start any other way (how that guard works).
No Docker? uv sync && MUNNIN_AUTH=off uv run python -m munnin does the same on 127.0.0.1:8200.
The default is token mode: every /mcp and /api call needs a bearer token from an OIDC issuer that binds tokens to this server (aud = https://<your-host>/mcp), and each person who signs in gets their own tenant. Set MUNNIN_PUBLIC_BASE_URL and MUNNIN_LOGTO_ENDPOINT, put it behind TLS, and paste https://<your-host>/mcp into claude.ai's connectors or claude mcp add. The issuer requirements, the one-identifier rule, and the prebuilt image are in How Is It Deployed.
Full architecture, the tool and endpoint surface, the data model, and known debts: docs/README.md.
Apache License 2.0 — Copyright 2026 Alviandi Widiasto. The control-files submodule (agent-memory-system) is licensed the same way, so a clone with submodules is one licence throughout.