slashhuang

slashhuang / openclaw-traceflow

Public

watch dog and metrics dashboard for openclaw

13
0
100% credibility
Found Mar 23, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

OpenClaw TraceFlow is a standalone web dashboard that monitors OpenClaw AI agent sessions, token usage, skills, latency, prompts, pricing, and logs by connecting to a running OpenClaw Gateway.

How It Works

1
📰 Discover a helpful monitor

You hear about OpenClaw TraceFlow, a friendly dashboard to watch your AI assistant's chats, costs, and performance.

2
📥 Get it set up quickly

Download and launch the monitor app on your computer with a simple one-click start.

3
🔗 Connect your AI helper

Use the easy setup wizard to link it to your running AI gateway, testing the connection in seconds.

4
📊 See everything at a glance

Open the colorful dashboard showing live chat stats, token spending, speed, and health checks.

5
🔍 Explore chats and savings

Click into conversations, track skills used, watch token alerts, or check model costs to stay in control.

✅ Run your AI smarter

You now have clear insights to optimize chats, cut costs, and keep everything running smoothly.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 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 openclaw-traceflow?

OpenClaw TraceFlow is a standalone observability dashboard acting as a watchdog for OpenClaw AI agents, tracking sessions, skill invocations, token usage with alerts, latency percentiles (P50/P95/P99), system prompts, and live logs via a long-lived WebSocket. Built in JavaScript with NestJS backend and React frontend (Vite + Ant Design), it complements OpenClaw's default console by pulling data from your gateway (default localhost:18789) without replacing it. Deploy with one PM2 command for English/Chinese UI and JSON HTTP APIs—perfect for monitoring like you github watch repositories for commits, PRs, issues, and releases.

Why is it gaining traction?

It stands out with operator-safe overviews (no full operator.read scope needed), dual-track token metrics (recorded vs log-estimated for stale indexes), and in-product scope hints explaining exactly what stats include/exclude—avoiding the "numbers without context" trap in other agent consoles. Developers hook on quick deploys, realtime log streaming via Socket.IO, and extras like model pricing overrides and skill-tool breakdowns, far beyond basic gateway status. Like github watch vs star, it actively monitors production drifts instead of passive bookmarking.

Who should use this?

OpenClaw gateway operators debugging high-token sessions or zombie skills in prod deployments. Teams tracking per-user agent usage, latency spikes, or prompt bloat without digging transcripts manually. AI devs evaluating costs across models, especially in multi-tenant setups like group chats or scheduled jobs.

Verdict

Grab it if you're running OpenClaw—solid docs and PM2 quickstart make it dead simple despite 13 stars and 1.0% credibility score signaling early maturity (no heavy tests yet). Watch this repo like a watchdog; it'll mature fast for serious agent fleets.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.