Background Process Manager

Background Process MCP provides AI assistants with the ability to start, stop, and monitor long-running command-line processes through a server-centric architecture built by Joel (waylaidwanderer). The implementation uses a standalone WebSocket server that manages child processes via node-pty, maintaining complete process state including output history, with both an MCP client for LLM integration and an optional terminal UI for human monitoring. It addresses the limitation that many AI agents like Google's Gemini CLI cannot manage background processes natively, offering tools for process lifecycle management, output retrieval with head/tail options, and real-time monitoring, making it valuable for agents that need to run servers, watchers, build processes, or other long-running tasks while maintaining visibility into their execution status.

Source

Repository: https://github.com/waylaidwanderer/background-process-mcp