The pmb Model Context Protocol (MCP) server provides seamless integration between AI assistants (such as Claude Desktop, Cursor, and Claude Code) and Local-first persistent memory for AI coding agents (Claude Code, Cursor, Codex) over MCP. Decisions, lessons and facts live in one SQLite file on your disk. Offline, multilingual. By connecting this server, your AI agent gains specialized capabilities to inspect, query, and automate tasks within the AI/LLM domain.
Once installed in Claude Desktop or Cursor, try asking your assistant:
Copy the config for your editor. Verify environment variables or setup notes in the README.
Add to claude_desktop_config.json:
{
"mcpServers": {
"pmb": {
"command": "uvx",
"args": [
"pmb"
]
}
}
}
Or install with pip: pip install pmb
<!-- mcp-name: io.github.oleksiijko/pmb-ai --> <div align="center"> <img src="https://raw.githubusercontent.com/oleksiijko/pmb/main/docs/assets/logo.png" width="100" alt="PMB logo"> <img src="https://raw.githubusercontent.com/oleksiijko/pmb/main/docs/assets/dashboard-map.png" width="760" alt="PMB dashboard - your project's memory as a live entity graph">