Arize Phoenix MCP

Arize AI's Phoenix, an open-source AI observability and evaluation platform, exposes a remote MCP endpoint that connects agents directly to a Phoenix instance. Lets assistants query OpenTelemetry traces and spans from LLM applications, inspect datasets and experiments, review evaluation results, and investigate latency or quality regressions.

{
  "mcpServers": {
    "arize-phoenix": {
      "url": "https://github.com/Arize-ai/phoenix",
      "env": {
        "PHOENIX_COLLECTOR_ENDPOINT": "<YOUR_PHOENIX_COLLECTOR_ENDPOINT>",
        "PHOENIX_API_KEY": "<YOUR_PHOENIX_API_KEY>"
      }
    }
  }
}

{
  "type": "object",
  "required": [
    "project_name"
  ],
  "properties": {
    "project_name": {
      "type": "string",
      "description": "Phoenix project name"
    },
    "filter_condition": {
      "type": "string",
      "description": "Optional filter expression over span attributes"
    }
  }
}

{
  "type": "object",
  "required": [
    "dataset_name"
  ],
  "properties": {
    "dataset_name": {
      "type": "string",
      "description": "Dataset name"
    }
  }
}