mcp-sync

A project's MCP servers are recorded in one doc — MCP.md at the repo root — and you mirror them into the three agents' config files. There is no script; you do this by following the steps below. It's safe to do by hand because these are small, project-local files: read one whole, change only its MCP section, write it back.

Scope is deliberately narrow: project-level only, three targets (Claude Code, opencode, Codex). No global/user configs, no other tools.

The source of truth — MCP.md

A markdown doc you (the agent) maintain at the project root. The structured server list lives in a fenced ```yaml block; prose around it carries notes (per-tool auth, etc.). See examples/example.MCP.md for the shape. Sketch:

# MCP servers for this project