# maestro-ai

Source: https://glimind.com/server/npm%2Fmaestro-ai
Registry id: `npm/maestro-ai`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "maestro-ai": {
      "command": "npx",
      "args": [
        "-y",
        "maestro-ai"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add maestro-ai -- npx -y maestro-ai
```

## Tools (8)

- `maestro_analyze`: Per-task routing insights, recommendations, learned-routing readiness.
- `maestro_ask`: Route and execute an LLM call with optional workflow, returning a routing report.
- `maestro_doctor`: Infrastructure diagnostics: process, port, /v1/models, env vars.
- `maestro_feedback`: Provide structured feedback on a routing decision.
- `maestro_probe`: Health-check each tier primary and fallback endpoints.
- `maestro_route`: Analyze task and return tier/model without making an LLM call. Always includes analysis, debug, probe, fallback_reason.
- `maestro_stats`: Telemetry summary: tier mix, escalation rate, latency, cost. Optionally include routing analysis.
- `maestro_workflow`: Multi-step orchestration (auto, critique, implement-test-fix, etc.).

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.