mcp_serverDeveloper Tools
OfficialCrawl4AI Model Context Protocol (MCP) Server
Works with:Crawl4AIClaude DesktopCursor
integrationDeveloper Toolsfree
Official Platform Integration Verified Integrity
Langfuse FastAPI LLM Observability Middleware
Trace nested model calls, token costs, and API latency with automatic Langfuse telemetry
What it does
Production pattern for instrumenting Python FastAPI backend services with Langfuse tracing. Wraps asynchronous route handlers to log prompt inputs, model responses, token counts, and execution latency into Langfuse dashboard traces.
Installation & Setup
Level: intermediate1. Install Langfuse SDK: `pip install langfuse`.
2. Set environment variables: `LANGFUSE_PUBLIC_KEY`, `LANGFUSE_SECRET_KEY`, `LANGFUSE_HOST`.
3. Decorate your LLM functions with `@observe()`.
Prerequisites: Python 3.9+, Langfuse Cloud account or self-hosted instance
Security & Privacy Notes
Data transmitted to your Langfuse host. Use self-hosted Langfuse if zero external telemetry is mandated.
Related AI Tools on AlsoClick
Leading AI platforms compatible with Langfuse FastAPI LLM Observability Middleware.