Qa Trace

QA Trace records your actions and monitors runtime errors as you explore a web application, then generates structured prompts you can send to an LLM to produce bug reports, test documentation, or full session reports.

Three prompt modes:
1. Steps to Reproduce for Ticket — generates concise step-by-step instructions focused on key actions and errors, ready for a bug ticket.
2. Document steps — produces neutral action-only documentation without error context.
3. Full Report of Actions — creates a comprehensive timeline of all tracked interactions and errors.

Error monitoring:
Detects JavaScript console errors, network failures (with request/response details), and UI errors via configurable CSS selectors. Captures screenshots for UI errors automatically.

Integrations:
Send generated prompts directly to OpenAI, DeepSeek, or any OpenAI-compatible LLM endpoint. Trigger webhooks to push data to external systems.

Privacy-first:
No data leaves your browser unless you explicitly send it. Tracking only runs on URLs you configured. Sensitive HTTP headers and body fields are automatically redacted. All stored data auto-expires after 12 hours.

Configurable: Set action and error limits, text length caps, URL privacy (strip query strings), output language, and more.