Skip to content
mcp/harbor
VS Code

ModelLens for VS Code

io.github.greatbeing/modellens

See which models actually work: probe, rank and recommend across OpenAI-compatible gateways.

client:VS Code transport:stdio runtime:pypi

Install ModelLens in VS Code

.vscode/mcp.json 

{
  "inputs": [
    {
      "type": "promptString",
      "id": "modellens-config",
      "description": "MODELLENS_CONFIG",
      "password": true
    },
    {
      "type": "promptString",
      "id": "modellens-data",
      "description": "MODELLENS_DATA",
      "password": true
    }
  ],
  "servers": {
    "modellens": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "modellens-mcp"
      ],
      "env": {
        "MODELLENS_CONFIG": "${input:modellens-config}",
        "MODELLENS_DATA": "${input:modellens-data}"
      }
    }
  }
}

VS Code prompts for each input the first time the server starts, so no secret is written to the file. VS Code docs

ModelLens in other clients

ModelLens MCP server