
by skynetcmd
m3 Memory is a local-first memory layer that lets AI coding assistants keep facts, decisions and conversation history across sessions and tools. Assistants reach it through MCP, and it also has a command-line interface.
PluginContact for pricing
Step 1: Install with pipx install m3-memory (or pip in a virtual environment).
Step 2: Run m3 setup so it detects your agents and connects them to m3.
Step 3: Run m3 doctor to check that everything is working.
Step 4: Tell your assistant something to remember, such as a project decision.
Step 5: In a new session, ask where you left off and the assistant can recall it.
No. m3 itself needs no cloud account or API key and works offline, though your AI assistant may have its own requirements.
The README names Claude Code, Codex, Cursor and Gemini CLI, connected through MCP.
On your own computer in a SQLite file by default. PostgreSQL is an alternative for sharing one store across machines.
Requirements
Inspeximus MCPInspeximus is tamper-evident long-term memory for AI agents, shipped as one zero-dependency Python file plus an MCP server. A corrected fact stays corrected, and erasures can be proven.