Langfuse MCP server

Endpoint: https://cloud.langfuse.com/api/public/mcp · Authentication: API key header

Reads and writes Langfuse project data such as prompts, traces, and observations. Requires the project public and secret key pair sent with HTTP Basic authentication.

How do I authenticate?

Create a scoped API key with the provider and configure Authorization in your client. Use the generated setup below; do not put the key in the endpoint URL.

Authentication and connection guide

Connection specification

Remote endpoint
https://cloud.langfuse.com/api/public/mcp
Browser access
Not yet measured
Declared MCP transport
Streamable HTTP
Live-validated MCP transport
Not yet verified — Latest validation did not verify a transport
Declared authentication
API key
Detected authentication
API key detected or retained from publisher evidence
Alternative endpoint
https://us.cloud.langfuse.com/api/public/mcp — Use for Langfuse US cloud projects (API key)
Protocol lifecycle
Not yet negotiated
Required header
Authorization — Base64 of the project public key and secret key joined by a colon: Basic <LANGFUSE_BASIC_AUTH>

Product documentation · Source repository · Official listing documentation

More AI Infrastructure MCP servers · Browse all MCP servers · Test this endpoint in the MCP Playground

Authorization setup

API key required

Use the publisher-documented API key route. Keep the credential in protected client storage and replace only the named placeholder at connection time.

Safe header template: Basic <LANGFUSE_BASIC_AUTH>. Replace only the named placeholder in protected client configuration.

Publisher documentation

Connect it

Live-validated endpoint: https://cloud.langfuse.com/api/public/mcp · Streamable HTTP. API key authentication uses the publisher-documented header name when available.

Agent prompt for Claude Code or Codex
Add the Langfuse MCP server to my MCP client configuration and verify that it works.

What it is: Reads and writes Langfuse project data such as prompts, traces, and observations. Requires the project public and secret key pair sent with HTTP Basic authentication.
Server URL: https://cloud.langfuse.com/api/public/mcp
Transport: Streamable HTTP (remote MCP server, no local install).
Authentication: API key. API key authentication uses the publisher-documented header name when available.
Tools: not yet observed by mcptest; tools/list will show them.
Reference: https://mcptest.io/servers/langfuse/

Steps:
1. Work out which MCP client I use (Claude Code, Codex, Cursor, VS Code or another) and add the server under the name "langfuse" in that client's MCP configuration, user scope unless I ask otherwise.
   Claude Code: claude mcp add --transport http --scope user --header 'Authorization: Basic '"${LANGFUSE_BASIC_AUTH}" 'langfuse' 'https://cloud.langfuse.com/api/public/mcp'
   Other clients: use the setup for that client on the reference page.
2. Never write a credential value into a file that could be committed; reference an environment variable or the client's secret store, and ask me for any value you need.
3. Verify: connect to the server and call tools/list. Confirm the tools above are listed, and report any sign-in step I still have to complete.
MCP client

Claude Code setup

Run in a terminal. This user-scoped entry is available across projects.

claude mcp add --transport http --scope user --header 'Authorization: Basic '"${LANGFUSE_BASIC_AUTH}" 'langfuse' 'https://cloud.langfuse.com/api/public/mcp'
  • Store LANGFUSE_BASIC_AUTH in your operating system keychain, secret manager, or protected environment; never commit the value. The Authorization header syntax is Basic <LANGFUSE_BASIC_AUTH>.

Codex CLI setup

Add to ~/.codex/config.toml, then securely set LANGFUSE_BASIC_AUTH_HEADER to the complete Authorization header value, including the required syntax Basic <credential>.

[mcp_servers.langfuse]
url = "https://cloud.langfuse.com/api/public/mcp"
env_http_headers = { "Authorization" = "LANGFUSE_BASIC_AUTH_HEADER" }
  • Store LANGFUSE_BASIC_AUTH_HEADER in your operating system keychain, secret manager, or protected environment; never commit the value. Set it to the complete Authorization header value, including the required syntax Basic <credential>.

Cursor setup

Add to .cursor/mcp.json for this project, or ~/.cursor/mcp.json for all projects.

{
  "mcpServers": {
    "langfuse": {
      "url": "https://cloud.langfuse.com/api/public/mcp",
      "headers": {
        "Authorization": "Basic ${env:LANGFUSE_BASIC_AUTH}"
      }
    }
  }
}
  • Store LANGFUSE_BASIC_AUTH in your operating system keychain, secret manager, or protected environment; never commit the value. The Authorization header syntax is Basic <LANGFUSE_BASIC_AUTH>.

VS Code setup

Add to .vscode/mcp.json, or run “MCP: Open User Configuration” for a private user-level entry.

{
  "servers": {
    "langfuse": {
      "type": "http",
      "url": "https://cloud.langfuse.com/api/public/mcp",
      "headers": {
        "Authorization": "Basic ${input:langfuse_basic_auth}"
      }
    }
  },
  "inputs": [
    {
      "type": "promptString",
      "id": "langfuse_basic_auth",
      "description": "LANGFUSE_BASIC_AUTH for Langfuse",
      "password": true
    }
  ]
}
  • VS Code requests LANGFUSE_BASIC_AUTH as a masked input; use a secret manager and do not put a default value in mcp.json. The Authorization syntax remains Basic <LANGFUSE_BASIC_AUTH>.

Claude setup

Add the endpoint above in Claude:

  • Customize > Connectors > Add custom connector
  • Team/Enterprise: Organization settings > Connectors > Add > Custom > Web

Claude Desktop setup

Add the endpoint above in Claude Desktop:

  • Remote: Customize > Connectors (same as claude.ai)
  • Local: Settings > Developer > Edit Config, or Settings > Extensions for .mcpb files
  • ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), local servers only

ChatGPT setup

Add the endpoint above in ChatGPT:

  • ChatGPT web: Plugins > Add custom MCP server, then configure authentication and accept the risk warning

Windsurf setup

Add the endpoint above in Windsurf:

  • ~/.config/devin/mcp_config.json (%APPDATA%\devin\mcp_config.json on Windows), serverUrl or url
  • Cascade panel > ... > MCPs > Open MCP config file

Gemini CLI setup

Add the endpoint above in Gemini CLI:

  • gemini mcp add --transport http <name> <url>
  • ~/.gemini/settings.json (user) or .gemini/settings.json (project), mcpServers

MCP SDK setup

Add the endpoint above in MCP SDK:

  • Programmatic: new Client(...).connect(new StreamableHTTPClientTransport(url))

Notes for every client

  • Alternative endpoint: https://us.cloud.langfuse.com/api/public/mcp — Use for Langfuse US cloud projects.

Client compatibility

ClientStatusWhy
Claude CodeSupportedClaude Code supports Streamable HTTP servers and API key headers.
CodexSupportedCodex supports Streamable HTTP servers and API key headers.
CursorSupportedCursor supports Streamable HTTP servers and API key headers.
VS CodeSupportedVS Code supports Streamable HTTP servers and API key headers.
ClaudePartialClaude supports API key headers only in limited availability. static_headers: beta, for a limited set of organizations.
Claude DesktopPartialClaude Desktop supports API key headers only in limited availability. static_headers: beta, for a limited set of organizations.
ChatGPTNot supportedChatGPT does not support API key headers. ChatGPT cannot present custom API keys.
WindsurfSupportedWindsurf supports Streamable HTTP servers and API key headers.
Gemini CLISupportedGemini CLI supports Streamable HTTP servers and API key headers.
MCP SDKSupportedMCP SDK supports Streamable HTTP servers and API key headers.

Computed from this server's validated transport, protocol and authentication against documented client support. Checks the catalog has not observed are not counted.

Similar servers

  • Hugging Face — Hugging Face's hosted MCP server explores models, datasets, Spaces, papers, jobs, and community AI tools. OAuth 2.1
  • Cloudflare AI Gateway — Searches AI Gateway request logs and reads prompt and response details for model traffic routed through Cloudflare. Requires Cloudflare OAuth. OAuth 2.1
  • Cloudflare AutoRAG — Lists and queries the AutoRAG (AI Search) retrieval instances in a Cloudflare account. Requires Cloudflare OAuth. OAuth 2.1
  • Mem0 — Adds, searches, updates, and deletes long-term memories stored in a Mem0 account so agents can persist context across sessions. Requires Mem0 OAuth. OAuth 2.1
  • Braintrust — Queries Braintrust logs, experiments, and datasets with SQL and manages prompts, evaluators, and evals. Supports OAuth or a Braintrust API key as a bearer token. OAuth 2.1
  • Weights & Biases — Queries W&B runs, run history, Weave traces and evaluations, artifacts, and the registry, and creates W&B reports. Requires a W&B API key as a bearer token. Bearer token

Guides for this server

Watch this server for breaking changes

Share and embed

Maintain Langfuse? Embed this badge in your README. It follows the latest catalog validation result and links to this server profile.

mcptest.io: auth required

Latest validation evidence

Validation status
Online when last tested
Validation checked at
Oct 26, 2026 at 11:06 AM
Validation detail
Authentication challenge at https://cloud.langfuse.com/api/public/mcp returned HTTP 401; endpoint was reachable but did not complete an MCP probe; authentication api-key

Tool inventory

Requires authorization; inventory not observed

The catalog validator lists tools only on servers that answer without credentials. Langfuse asks for an API key first, so its tools, resources and prompts were not observed. Connect in the Playground with your own credentials to list them.