BrowserTrace is an open-source project for debugging AI browser-agent failures. The project should be useful to people working across different agent stacks, companies, and experience levels.
- Be respectful and constructive in issues, discussions, reviews, and pull requests.
- For small docs or example fixes, use the First PR Recipe to keep the first contribution small and reviewable.
- Keep feedback technical and specific. Explain the failing workflow, agent stack, trace field, or integration problem you are discussing.
- When reporting stack-specific debugging feedback, check the closest guide
first:
- Browser Use guide: https://aaronlab.github.io/browsertrace/browser-use-debugging.html
- Stagehand guide: https://aaronlab.github.io/browsertrace/stagehand-debugging.html
- Skyvern guide: https://aaronlab.github.io/browsertrace/skyvern-debugging.html
- Playwright + LLM guide: https://aaronlab.github.io/browsertrace/playwright-llm-debugging.html
- Computer-use guide: https://aaronlab.github.io/browsertrace/computer-use-agent-debugging.html
- Assume trace data may be sensitive. Do not post private screenshots, prompts, model outputs, URLs, cookies, tokens, customer data, or proprietary workflows.
- Use
browsertrace export <run_id> --public -o public.htmlbefore sharing a real trace in public. - Respect maintainers' scope decisions. BrowserTrace is local-first and focused on browser-agent failure debugging.
- Harassment, threats, personal attacks, or discriminatory language.
- Posting secrets, private customer data, or non-consensual screenshots.
- Spamming issues, discussions, or external communities with repeated promotion.
- Asking for stars, upvotes, vote swaps, fake engagement, or automated promotion.
- Derailing technical discussions with unrelated arguments.
For moderation or conduct concerns, contact the maintainer privately when
possible. For security-sensitive reports, follow SECURITY.md instead of
opening a public issue.
Maintainers may edit or remove comments, lock threads, close issues, or block accounts when behavior harms the project or its users.