doneyli

doneyli / claude-code-langfuse-template

Public

Self-hosted Langfuse for Claude Code session observability

59
17
100% credibility
Found Feb 04, 2026 at 36 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A self-hosted template that runs a local dashboard to capture, group, and visualize all interactions from Claude Code conversations including prompts, responses, and tool calls.

How It Works

1
📖 Discover the chat tracker

You read a blog post about a simple way to record and review all your conversations with your AI coding assistant Claude Code.

2
🛠️ Check your setup

You make sure your computer has the everyday tools like a web container runner and recent Python ready, just like installing common apps.

3
📥 Download and secure it

You grab the ready-made files and create your own private login details with a quick helper script.

4
🚀 Launch your dashboard

With one simple command, you start up your personal viewing dashboard that runs safely on your own computer.

5
🔗 Connect to Claude

You link the tracker to your Claude Code so it automatically captures every chat turn without interrupting your work.

6
💬 Start chatting

You open Claude Code, have a normal conversation with your AI, and watch the magic happen in real-time.

👀 See everything clearly

Now you have a beautiful dashboard showing every prompt, response, tool use, and session grouped neatly for easy review and learning.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 36 to 59 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is claude-code-langfuse-template?

This template spins up a self-hosted Langfuse instance via Docker Compose to monitor your Claude Code sessions, capturing prompts, responses, tool calls, and full conversation threads locally. It solves the problem of opaque AI interactions by grouping turns into traces viewable at localhost:3050, with control over data retention and no cloud dependency—ideal for langfuse self hosted docker setups. Setup takes under 5 minutes: generate env vars with a shell script, start services, and install a Python hook into Claude Code.

Why is it gaining traction?

Unlike cloud Langfuse, this offers full self-hosted control with Docker Compose, including PostgreSQL, ClickHouse analytics, and MinIO storage, letting you tweak langfuse self hosted data retention or ports easily. Developers dig the incremental processing that avoids re-parsing old transcripts, real-time session grouping, and per-project opt-outs via simple JSON configs. It's a turnkey langfuse self hosted guide for Claude users wary of SaaS telemetry.

Who should use this?

Claude Code power users debugging tool chains or analyzing response patterns in personal projects. Devs running self-hosted setups like GitHub runners in Docker who want LLM observability without external services. Teams evaluating langfuse self hosted vs cloud for enterprise privacy needs.

Verdict

Grab it if you're deep into Claude Code and need local traces—docs are thorough, setup scripts solid despite 53 stars and 1.0% credibility score signaling early maturity. Skip for production without your own hardening; it's a personal project with no guaranteed support.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.