Model Shunt MCP Server

by yasmanycastillo Your server? Claim it

local only

Model Shunt runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 24 Sep 2026.

Model routing for AI agents: delegate bulk reads & boilerplate to cheap worker models.

—

Installs / week

pypi · model-shunt

Stars

0 open issues

24 Sep 2026

Last commit

no recent releases

MIT

License

Python

Nothing serious here today

Today is the operative word: we check Model Shunt every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.

Connect this server

This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.

Claude Code Claude Desktop Codex Cursor VS Code

run in your terminal

claude mcp add model-shunt -- uvx model-shunt

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "model-shunt": {
      "args": [
        "model-shunt"
      ],
      "command": "uvx"
    }
  }
}

~/.codex/config.toml

[mcp_servers.model-shunt]
command = "uvx"
args = ["model-shunt"]

.cursor/mcp.json

{
  "mcpServers": {
    "model-shunt": {
      "args": [
        "model-shunt"
      ],
      "command": "uvx"
    }
  }
}

.vscode/mcp.json

{
  "mcpServers": {
    "model-shunt": {
      "args": [
        "model-shunt"
      ],
      "command": "uvx"
    }
  }
}

Alternatives to Model Shunt

same job, measured the same way

MCP Bifrost

by fixembcn

Delegate bulk code edits to a cheap worker model. Gated, atomic. PHP and Python.

176 installs/wk local only

Callable

by ccstt-dev

API for AI agents to delegate tasks to real humans.

3 tools answering

Cheaplane

by millennialdreamer

Offload grunt work from your premium agent to cheap models — auto-routing and billing isolation.

80 installs/wk local only

Routing Services Courier

by carlos-vlab

Pay-per-call ($0.01 USDC) model-routing for AI coding agents: which LLM to call, cost vs. quality.

1 tools answering

Qodercli MCP

by cantbeblank96

MCP server mode for Qoder CLI — delegate coding tasks to local Qoder agents from any MCP client.

81 installs/wk local only

Sub Agents MCP

by shinpr

MCP server for delegating tasks to specialized AI sub-agents across coding tools

110 installs/wk local only

Word MCP Server (FsWordDsl)

by marknicholls

Word tools for AI agents: read/write, decompile to F#/C#/XML/JSON, or build from XML/JSON.

local only

Agentorist

by agentorist

Booking gateway for AI agents — discover events, movies & hotels, hand off to partner checkout.

7 tools answering

Model Shunt — questions

Answers built from our own checks of this server.

Why is there no uptime for Model Shunt?

Model Shunt runs on your own machine over stdio — there is no network address to reach, so uptime cannot be measured for it by anyone. What can be measured is adoption: the pypi package model-shunt.

How do I connect Model Shunt?

Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It runs locally, so the command pulls model-shunt straight from pypi; nothing to host, nothing to sign up for.

Is Model Shunt open source?

Yes — it is published under the MIT licence, written in Python and 0 stars on GitHub. The source link is on this page, so you can read exactly what it does with your data before you connect it.