Run a free break-risk scan on a script you already bought.
Paste an n8n/Make export, a webhook handler, or any automation snippet. We run real client-side heuristic checks — hardcoded secrets, code-exec, brittle selectors, fragile integrations, missing error handling — and hand you a break-risk scorecard. No upload, no account.
Heuristic, illustrative pre-check — pattern-matching only, not the full certification scan and not a security guarantee. Your code is analysed locally in your browser and is never transmitted or stored.
Your scorecard appears here
Paste a script on the left and run the scan. Try the sample to see a deliberately fragile one graded.
Lower break-risk is safer. The score weighs external dependencies, secrets exposure, execution surfaces, brittleness signals and resilience patterns. It's a preview of the model behind every Certified badge — not a substitute for the full scan.
Get a full break-risk audit of one listing — free.
Like what the preview showed you? We'll run the complete certification scan on one script you're thinking of buying or selling, and send the full scorecard. No charge for the first one.
- Full static + dynamic scan, not just the heuristic preview
- Dependency watchlist scoring + concrete mitigations
- A bond + SLA quote if you decide to certify it
Scans run · 0 on this browser
Quick questions about the free tool.
Does my code get uploaded anywhere?
No. The break-risk scan runs entirely in your browser with JavaScript pattern-matching. Your script is never sent to a server. The only thing stored — and only if you submit the form — is the numeric score and your contact details, in this browser's localStorage for the demo.
How accurate is the heuristic score?
It's a fast, directional read — great for catching obvious red flags (hardcoded secrets, exec paths, brittle selectors) and rewarding good patterns (error handling, retries, env config). The full certification scan goes far deeper, including dependency watchlist scoring. Treat the preview as a smoke test, not a verdict.
What counts toward break-risk?
Distinct external endpoints (each is a break surface), exposed secrets, dynamic code execution and injection surfaces, brittle DOM scraping, versioned vendor APIs, and fixed waits without retries — all push the score up. Try/catch, retry/backoff, environment-based config and idempotency patterns pull it down.