Ambit: every MCP server your AI agents have, and what to add next
Ambit reads the MCP configs of Claude Code, Cursor, Codex CLI and 13 more agents on your machine and draws what they add up to as one local map: every server, which ones work, and what one more step would open. What you, your agents and your machines can jointly do is your ambit, and the map shows where it ends and how to widen it.
What it answers
- What is worth setting up next, ranked by what it would open and the attention it would save.
- What compound abilities emerge when two tools are combined.
- Every MCP server each agent has, which of them pass their checks, and what may run without asking.
- What is configured but failing, and what breaks downstream if a model, MCP server, tool or credential goes away.
You ask from the terminal. Your agents ask over MCP: Ambit is itself an MCP server, with a small profile for an agent's context and the full roster behind it. An agent can propose a configuration change; applying one takes an approval, made outside MCP. Everything stays in a local SQLite file.
To run it on your own configs: npx ambit-cli, with Node 22.18 or newer. The interactive map and the live demo need JavaScript. These pages say what it does without it:
- Guide: install, the CLI, and connecting it to Claude Code or OpenCode
- FAQ: what it reads, and what leaves your machine
- Where each AI agent keeps its MCP config: Claude Code, Cursor, Codex CLI, Goose, Kiro and more
- What stops when an MCP server or credential expires: finding single points of failure
- Auditing an agent's MCP servers: declared, working, permitted, and used
- Reference: the model and every MCP tool
- Ambit and Jev: typed judgment on the capability map
- All documentation
- Plain-text guide for agents, including direct links to demo views and contributor commands
- Source on GitHub, MIT licensed