langfuse-mcp

概述

langfuse-mcp服务器通过模型上下文协议(MCP)提供对Langfuse可观测性数据的全面访问。它通过37个工具覆盖多个类别，包括追踪、观察、会话、异常、提示词、数据集、注释队列和评分。这使AI代理能够直接查询生产遥测数据，回答关于性能故障、用户会话、提示词版本和数据集注释的关键问题。该服务器由积极维护，具有最近的提交记录，并提供只读模式和选择性工具加载以高效使用token。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• fetch_traces

  Fetch Langfuse trace data for debugging and analysis

  试试:Find traces with errors in the last 24 hours试试:Show me slow traces with high latency
• fetch_trace

  Get details for a specific Langfuse trace

  试试:Show me details for trace abc123试试:What happened in trace 456def?
• fetch_observations

  Retrieve observations from Langfuse traces

  试试:Show all observations from yesterday试试:Find generations with unusual outputs
• find_exceptions

  Find exceptions in Langfuse traces

  试试:Show me all exceptions in the last week试试:Find exceptions related to authentication
• get_session_details

  Get details for a specific Langfuse session

  试试:Show me details for session xyz789试试:What happened in this user's session?
• list_prompts

  List available prompts in Langfuse

  试试:Show me all prompts试试:Find prompts labeled as production
• get_prompt

  Retrieve a specific prompt from Langfuse

  试试:Show me the latest version of my prompt试试:Get prompt with ID prompt123
• list_datasets

  List available datasets in Langfuse

  试试:Show me all datasets试试:Find datasets for testing purposes
• create_dataset_item

  Create or update a dataset item in Langfuse

  试试:Add a new item to my test dataset试试:Update existing item with ID item_123
• list_annotation_queues

  List annotation queues in Langfuse

  试试:Show me all annotation queues试试:Find queues with pending items
• list_scores_v2

  List scores in Langfuse v2

  试试:Show me all scores for my application试试:Find high-scoring generations
• get_data_schema

  Get the data schema for Langfuse

  试试:Show me the Langfuse data schema试试:What fields are available in traces?

可对比工具

安装

{
  "mcpServers": {
    "langfuse": {
      "command": "uvx",
      "args": ["langfuse-mcp"],
      "env": {
        "LANGFUSE_PUBLIC_KEY": "pk-...",
        "LANGFUSE_SECRET_KEY": "sk-...",
        "LANGFUSE_HOST": "https://cloud.langfuse.com"
      }
    }
  }
}

# Claude Code
code mcp add \
  -e LANGFUSE_PUBLIC_KEY=pk-... \
  -e LANGFUSE_SECRET_KEY=sk-... \
  -e LANGFUSE_HOST=https://cloud.langfuse.com \
  langfuse -- uvx langfuse-mcp

# Codex CLI
codex mcp add langfuse \
  --env LANGFUSE_PUBLIC_KEY=pk-... \
  --env LANGFUSE_SECRET_KEY=sk-... \
  --env LANGFUSE_HOST=https://cloud.langfuse.com \
  -- uvx langfuse-mcp

FAQ

这与官方的Langfuse MCP服务器相比如何？

这个服务器提供完整的可观测性工具包，包括追踪、观察、会话、异常、提示词、数据集、注释队列和评分。官方MCP只关注提示词管理，使这个解决方案在调试和分析代理行为方面更加全面。

我可以限制加载的工具以减少token使用吗？

是的，使用选择性工具加载 `langfuse-mcp --tools traces,prompts` 只加载特定的工具组。可用的组包括追踪、观察、会话、异常、提示词、数据集、注释队列、评分和模式。

langfuse-mcp 对比