Arize Phoenix MCP
LLM observability platform exposing prompts, projects, traces, spans, sessions, datasets, and experiments through MCP. Published to npm as @arizeai/phoenix-mcp, current 4.0.8 (2026-04-29). 9,496 stars on parent monorepo, Elastic License 2.0.
INSTALL THIS SERVER
{
"mcpServers": {
"phoenix": {
"command": "npx",
"args": [
"-y",
"@arizeai/phoenix-mcp@latest",
"--baseUrl",
"https://my-phoenix.com",
"--apiKey",
"<your-api-key>"
]
}
}
}
{
"mcpServers": {
"phoenix": {
"command": "npx",
"args": [
"-y",
"@arizeai/phoenix-mcp@latest",
"--baseUrl",
"https://my-phoenix.com",
"--apiKey",
"<your-api-key>"
]
}
}
}
{
"mcpServers": {
"phoenix": {
"command": "npx",
"args": [
"-y",
"@arizeai/phoenix-mcp@latest",
"--baseUrl",
"https://my-phoenix.com",
"--apiKey",
"<your-api-key>"
]
}
}
}
{
"mcpServers": {
"phoenix": {
"command": "npx",
"args": [
"-y",
"@arizeai/phoenix-mcp@latest",
"--baseUrl",
"https://my-phoenix.com",
"--apiKey",
"<your-api-key>"
]
}
}
}
{
"mcpServers": {
"phoenix": {
"command": "npx",
"args": [
"-y",
"@arizeai/phoenix-mcp@latest",
"--baseUrl",
"https://my-phoenix.com",
"--apiKey",
"<your-api-key>"
]
}
}
}
8 SELECTED TOOLS
OUR ASSESSMENT
Phoenix is the open-source observability and evaluation platform from Arize. The @arizeai/phoenix-mcp package surfaces the platform main workflows over MCP. Tools cover prompt management (list-prompts, upsert-prompt, version tags), projects, traces, spans, sessions, annotation configs, datasets (with example management), and experiments. Connects to a self-hosted Phoenix instance or Phoenix Cloud via --baseUrl and --apiKey. The package is one of several MCP integrations in the Phoenix monorepo.
- 100 commits on main in the last 30 days on the parent monorepo.
- 9,496 GitHub stars.
- Published to npm as @arizeai/phoenix-mcp with semantic versioning (current 4.0.8, released 2026-04-29).
- Tool surface covers prompts, projects, traces, spans, sessions, annotation configs, datasets, and experiments.
- Phoenix is widely adopted in the LLM observability space; this MCP plugs directly into existing Phoenix deployments.
- One-click Cursor install link in the README.
- Elastic License 2.0 is source-available, distinct from a permissive open-source licence; review ELv2 terms for downstream commercial use.
- Phoenix documentation search ships as a separate MCP server; the operator picks one or both.
- Requires a running Phoenix instance (self-hosted or Cloud) to be useful.
- API key handling: the MCP needs a Phoenix API key, which carries the same access as the dashboard user it represents.
The MCP carries Phoenix API key permissions through to the agent. Scope the API key to read-only when the agent does diagnostic work; reserve write-capable keys for workflows where prompt or dataset mutation is desired. Self-hosted Phoenix instances inside private networks reduce attack surface compared with Phoenix Cloud, at the cost of operational overhead. The Elastic License 2.0 permits production use; it restricts offering Phoenix as a competing managed service to third parties.
Teams running Phoenix as their LLM trace and evaluation surface who want agents to query traces, manage prompts, and inspect experiments through MCP; prompt-management workflows where the agent versions, tags, and retrieves prompts directly from Phoenix; dataset and experiment management inside Cursor, Claude Code, or VS Code.
TECHNICAL DETAILS
RECORDED ADOPTION METRICS
Saved directory snapshots. Record last checked MAY 1, 2026. The review date above does not refresh these figures.
// Recorded note9,496 stars on the parent Arize-ai/phoenix monorepo. 100 commits on main in the last 30 days.
// Recorded noteSecond-ranked in ai-ml on observability scope and adoption.
SOURCES & RECORDED CHECKS
These are the directory listings and check details held in this record. Client compatibility, directory listings and installation testing are separate signals. Recorded client names describe the scope of a check; they do not establish a security assessment or a recurring test schedule.
The same server, 5 different lenses. We reconcile these signals into our editorial score, which is why our number sometimes diverges from a directory-aggregate star count.
| Source | Their rating | Their star count | Their downloads | Last synced |
|---|---|---|---|---|
| AutomationSwitch This page | 4.5editorial | 9,496 | — | MAY 1, 2026 |
| PulseMCP | — unrated | unavailable | unavailable | MAY 1, 2026 |
| MCP.so | — unrated | unavailable | unavailable | MAY 1, 2026 |
| Glama | — unrated | unavailable | unavailable | MAY 1, 2026 |
| Smithery | — unrated | unavailable | unavailable | MAY 1, 2026 |
| Official MCP Registry | — unrated | unavailable | unavailable | MAY 1, 2026 |
// Counts are directory-reported; we don't adjust them. Discrepancies usually come from different snapshot times or star-caching.
OTHER AI / ML MCP SERVERS
Cognee MCP
Knowledge graph plus vector memory engine for AI agents, exposed as an MCP server with V2 session-aware memory tools (remember, recall, forget, improve) and classic V1 ingestion pipelines (cognify, codify). Three transports: stdio, SSE, Streamable HTTP. 16,965 GitHub stars, Apache-2.0.
HeyGen Hyperframes
Write HTML, render video. Built for agents. HeyGen's framework that turns HTML templates into video output via MCP. 16,400 GitHub stars and 100 commits on main in the last 30 days.
Codebase Memory MCP
High-performance code intelligence MCP server for AI coding agents. Indexes a codebase into a queryable knowledge graph in milliseconds, with 14 MCP tools spanning structural search, call-chain tracing, impact analysis, dead-code detection, and Cypher queries. Single static C binary, 66 languages via tree-sitter, zero runtime dependencies.
Qdrant MCP Server
Official Qdrant vector database MCP server. Acts as a semantic memory layer on top of Qdrant: store information with metadata, retrieve via similarity search. Two tools, very small surface area, exceptionally maintained by the Qdrant team. Configurable embedding provider (fastembed default), supports remote and local Qdrant clusters.
Mastra
TypeScript framework for building AI-powered applications and agents, with built-in MCP server support. From the team behind Gatsby. 23,680 GitHub stars and 100 commits on main in the last 30 days.
Amazon Bedrock AgentCore MCP
Official AWS Labs MCP server for Amazon Bedrock AgentCore: agent runtime, memory, gateway, identity, and observability. Tools fetch curated AgentCore documentation and surface deployment guides for runtime, memory, and gateway resources. Apache-2.0 within awslabs/mcp monorepo (8,924 parent stars).
DISCUSS YOUR
MCP REQUIREMENTS.
Evaluating a server, scoping an internal deployment, or working out whether MCP is the right fit at all. Start the conversation and we will point you at the right piece of the ecosystem.