🔍
Live in productionProductHigh complexity

TraceLens

Runtime truth engine for web applications — causality over metrics, AI-queryable.

< 1 ms
Browser SDK overhead
< 2 MB
Memory footprint
< 100 ms
API ingest latency
MIT (self-host)
License

What is TraceLens?

TraceLens transforms application debugging by focusing on causality rather than metrics. Instead of telling you "your app is slow," it builds real-time dependency graphs from browser and server runtime signals to show you exactly why it's slow.

It is self-hosted (your data never leaves your infrastructure), open-source under MIT, and integrates with AI coding assistants through the Model Context Protocol — so you can ask Claude or Kiro questions like "what are my app's current performance bottlenecks?" and get precise, grounded answers.

The most interesting design decision: TraceLens monitors its own dashboard performance in real time, so you experience the platform's capabilities before writing a single line of integration code.

Who is it for?

  • Developers using AI coding assistants (Claude, Kiro, Cursor, Windsurf) who want grounded performance context
  • Vibe-coding teams that don't want to wire up DataDog / New Relic just to see what's slow
  • Teams who care about causality (why is this slow?) more than dashboards (here are 50 metrics)
  • Anyone shipping browser apps where data privacy matters — TraceLens stays self-hosted

What does it do?

  • Real-time dependency graph from browser + server runtime signals
  • Critical-path detection (which exact call is blocking the response)
  • Web Vitals + automatic performance instrumentation out of the box
  • AI-queryable via MCP server — natural-language perf questions
  • Self-monitoring meta-observability (the dashboard monitors itself)
  • 5+ install modes: standard, quick, demo, enhanced, custom-port
  • Localhost-first deployment — no cloud dependency

How is it built?

100% TypeScriptBrowser SDKNode.js serverMCP serverReal-time graphsSelf-hosted

What makes it interesting?

  • Builds real causality graphs, not just metric dashboards
  • < 1 ms browser SDK overhead — you don't pay for the observability
  • Reduces AI debugging cost by ~80% via precise MCP context
  • Self-monitors the dashboard so you see it working before integrating
  • MIT licensed and fully self-hostable

Want something like this for your business?

We adapt the patterns above to your stack on every engagement. The 15-minute discovery call is free — you leave with a plan regardless.