Skip to content
mcp/harbor
Grok

model shunt for Grok

io.github.yasmanycastillo/model-shunt

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

client:Grok transport:stdio runtime:pypi

Install model shunt in Grok

.mcp.json (in your project root) 

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

Grok Build reads Claude Code's .mcp.json and Claude Desktop's config unchanged, so a server you have already set up there needs nothing new. The Grok web app and X integration do not take custom servers. Grok docs

model shunt in other clients

model shunt MCP server