TraceTriage

TraceTriage is a web app + CLI that pinpoints why API tests fail intermittently across environments. Instead of just reporting “expected 200, got 500,” it captures request/response diffs, timing, headers, auth context, and upstream dependency signals, then clusters failures into likely causes (rate limits, clock skew, schema drift, bad test data, nondeterministic mocks). It integrates with CI (GitHub Actions, GitLab CI, Jenkins) and popular test runners (Postman/Newman, Playwright API, pytest, Jest). The MVP focuses on collecting rich traces from failed runs, deduplicating noise, and generating a short “next steps” checklist per failure cluster. This is not a magic AI oracle: it’s a disciplined failure forensics tool that saves hours of reruns and guesswork. Teams can export artifacts for bug tickets and compare failures across branches and deployments.