| name | playwright-reporting |
| description | Generate non-interactive Playwright evidence reports that replace Playwright UI mode and HTML report browsing in OpenCaw sessions. |
When to use
Use when the user asks to inspect Playwright results, summarize browser test evidence, review screenshots/traces/videos, or produce a report in an environment where interactive Playwright UI tools are unavailable.
Typical triggers:
- "show the Playwright report"
- "summarize test results"
- "index screenshots and traces"
- "create a report from Playwright artifacts"
- "UI mode will not work"
Output
Produce one or more host-owned Markdown reports:
- test run summary from
test-results/results.json
- artifact index for screenshots, traces, videos, logs, HTML reports, JSON/XML reports, and snapshots
- browser discovery report from
.playwright-cli/
- bundled evidence report linking the generated reports
Notes
- Prefer
commands/playwright-evidence-report.sh for a one-command report bundle.
- Use
commands/playwright-report-summary.sh when JSON reporter output exists.
- Use
commands/playwright-artifact-index.sh to replace browsing report/artifact folders manually.
- Use
commands/playwright-discovery-report.sh to summarize browser discovery sessions.
- Reports should be written to the host
.ai/reports/ directory by default.
- Do not open Playwright UI mode or start the HTML report server unless the user explicitly asks for an interactive local view.