hyperroute-mcp

The official Model Context Protocol server for HyperRoute.

HyperRoute is a router for AI agents. Give it a task and it picks the best external tool for that task — measured, not advertised — then runs the tool for you with your own key held server-side, and learns from how it went. This MCP server is how a coordinator agent (Claude Code, Codex, Goose, Cursor, LangGraph, …) drives it:

recommend → onboard a key → execute the tool server-side → report the outcome

It talks to the router only over its public HTTP API and holds no product logic of its own.

Why route at all

An agent with 100 tools bolted on has a context problem and a quality problem. HyperRoute replaces both with one verb: your agent learns recommend, and HyperRoute decides which of hundreds of tools actually answers this task, whether you can already do it better yourself, and what it will cost.

• Measured, not advertised. Every capability score is backed by real graded probes you can inspect (describe(tool_id, ["evidence"])).
• Your keys never leave the server. You connect a key once; HyperRoute runs the tool with it and returns only the result. The key is never sent to your agent, never logged.
• It tells you when NOT to route. If nothing beats what your coordinator already does, the verdict is use_native — do it yourself. That only works if the server knows which coordinator it runs inside; see Your coordinator.

Install

Tell your coding agent:

Run uvx hyperroute-mcp@latest install and follow it.

It registers HyperRoute with your agent, then asks you before changing anything else. Needs uv.

By hand. Claude Code:

claude mcp add --scope user hyperroute -- uvx hyperroute-mcp@latest

then ask the agent to "finish HyperRoute setup". Any other MCP client:

{ "mcpServers": { "hyperroute": { "command": "uvx", "args": ["hyperroute-mcp@latest"] } } }

Setup, checks and upgrades

• finish_setup shows what it would add for your agent (in Claude Code: hooks that make the agent and its subagents route through HyperRoute) and applies it only when you agree.
• check_setup checks versions and the installed pieces, and upgrades when you agree.
• remove_setup (or hyperroute-mcp install --remove) undoes it.

Authenticate once

recommend and browsing are public — no account. Connecting keys and running tools need one. Mint a personal access token at hyperroute.io and hand it to the use_token tool; your password never enters the conversation. The login is cached on disk and restored in every new session until the router invalidates it.

Your coordinator

HyperRoute compares external tools against what your coordinator can already do. The server reads which coordinator launched it and declares it for you; session_info shows what it resolved.

Tools

A ranking looks like:

session: s-6d6c5a95f9f84d9a
verdict: interpose

  tool              name                        price  use        why
→ opencitations     OpenCitations Index         free   ready      highest-ranked: capability 0.81 …
  semantic_scholar  Semantic Scholar Graph API  free   needs_key  lower capability (0.75 vs 0.81).

confidence: med (on the pick)
act: execute('opencitations', <query>)

use: ready (run it) · needs_key (connect a key first) · native (the agent does it itself) · own (your own tool) · soon (not runnable yet).

Development

pip install -e ".[dev]"
pytest
ruff check .

The suite is fully offline — the router is faked, so no network and no real account are touched.

Environment variables

None are needed against the hosted router. For development and self-hosting:

License

MIT — see LICENSE.