Skip to content
mcp/harbor
VS Code

modelcostsaver for VS Code

io.github.sachinuppal/modelcostsaver

Offline MCP: predict LLM call cost and pick the cheapest capable model. No keys, no network.

client:VS Code transport:stdio runtime:npm

Install modelcostsaver in VS Code

.vscode/mcp.json 

{
  "servers": {
    "modelcostsaver": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@workswarm/modelcostsaver"
      ]
    }
  }
}

Requires agent mode. MCP: Open User Configuration puts the same block in every workspace. VS Code docs

modelcostsaver in other clients

modelcostsaver MCP server