# Arize Phoenix

> Use this tool when you need to manage and optimize large language models (LLMs) across multiple providers, or when building and evaluating LLM applications requires unified prompt management, dataset exploration, and experiment visualization. The Arize Phoenix MCP Server provides a unified interface for AI assistants to interact with the Phoenix platform, accepting inputs such as prompts and datasets and producing outputs like experiment results and visualizations. It is particularly useful in contexts where teams need to leverage observability features to improve LLM performance and reliability.

Canonical page: https://skillsregistry.net/skills/arize-phoenix  
JSON: https://api.skillsregistry.net/v1/skills/arize-phoenix

## Description

Phoenix MCP Server provides a unified interface to Arize Phoenix's capabilities through the Model Context Protocol. Developed by Arize AI, this TypeScript implementation enables AI assistants to manage prompts, explore datasets, and run experiments against the Phoenix platform. The server exposes tools for creating and iterating on prompts across different LLM providers (OpenAI, Anthropic, Google), working with evaluation datasets, and visualizing experiment results, making it particularly valuable for teams building and evaluating LLM applications who want to leverage Phoenix's observability features through AI assistants.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-05-27

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/arize-phoenix)
- **Repository:** <https://github.com/arize-ai/phoenix/tree/HEAD/js/packages/phoenix-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "arize-phoenix"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/arize-phoenix` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/arize-phoenix/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt