Goal
Add one short sentence to the Browser Use guide explaining when the local compare API is more useful than the CLI output.
Context: browsertrace compare <failed_run_id> <success_run_id> is good for humans in the terminal. The local API endpoint is better when a script, dashboard, or local UI wants the same first-divergence payload as JSON:
GET /api/compare/<failed_run_id>/<success_run_id>
Scope
Please keep this small:
- Update
docs/browser-use-debugging.html in the Compare failed and successful Browser Use runs section.
- Add one sentence near the existing curl example saying the endpoint is useful for local dashboards, scripts, or automation preflight checks that need the same first-divergence JSON payload.
- Update
tests/test_metadata.py so the Browser Use guide metadata test covers the new sentence.
Out of scope
- Do not change server behavior or API response shape.
- Do not add a UI diff view.
- Do not edit unrelated docs.
Verification
Please run:
uv run --python 3.11 --extra dev pytest tests/test_metadata.py -q
git diff --check
Comment here before starting so I can mark the issue as claimed.
Goal
Add one short sentence to the Browser Use guide explaining when the local compare API is more useful than the CLI output.
Context:
browsertrace compare <failed_run_id> <success_run_id>is good for humans in the terminal. The local API endpoint is better when a script, dashboard, or local UI wants the same first-divergence payload as JSON:Scope
Please keep this small:
docs/browser-use-debugging.htmlin theCompare failed and successful Browser Use runssection.tests/test_metadata.pyso the Browser Use guide metadata test covers the new sentence.Out of scope
Verification
Please run:
Comment here before starting so I can mark the issue as claimed.