Description
.github/workflows/docs-noob-tester.md instructs the agent (lines 141 and 155) to use playwright-cli browser_navigate --url "..." and playwright-cli browser_take_screenshot --filename .... The actual installed playwright-cli tool does not have these commands — it uses open/goto for navigation and screenshot for captures, per .claude/skills/playwright-cli/SKILL.md's command reference (playwright-cli open (example.com/redacted), playwright-cli goto (url), playwright-cli screenshot). As a result, every run of the Documentation Noob Tester workflow hits a tooling mismatch before falling back to manual open/goto/screenshot` calls, as confirmed by the 2026-10-01 Documentation Noob Test Report (discussion #64676).
Expected Impact
Removes a reproducible tooling error on every run of this workflow, so the agent uses the correct playwright-cli commands from the first attempt instead of discovering the mismatch at runtime.
Acceptance Criteria
Suggested Agent
Developer (workflow markdown change + recompile)
Estimated Effort
Quick (< 1 hour)
Data Source
DeepReport Intelligence Briefing, 2026-10-01 ~06:44Z cycle, sourced from Documentation Noob Test Report discussion #64676 (2026-10-01).
Generated by 🔬 Deep Report · claude · agent · 130.1 AIC · ⌖ 8.15 AIC · ⊞ 7.3K · ◷
Description
.github/workflows/docs-noob-tester.mdinstructs the agent (lines 141 and 155) to useplaywright-cli browser_navigate --url "..."andplaywright-cli browser_take_screenshot --filename .... The actual installedplaywright-clitool does not have these commands — it usesopen/gotofor navigation andscreenshotfor captures, per.claude/skills/playwright-cli/SKILL.md's command reference (playwright-cli open (example.com/redacted),playwright-cli goto (url),playwright-cli screenshot). As a result, every run of the Documentation Noob Tester workflow hits a tooling mismatch before falling back to manualopen/goto/screenshot` calls, as confirmed by the 2026-10-01 Documentation Noob Test Report (discussion #64676).Expected Impact
Removes a reproducible tooling error on every run of this workflow, so the agent uses the correct
playwright-clicommands from the first attempt instead of discovering the mismatch at runtime.Acceptance Criteria
docs-noob-tester.mdlines ~141 and ~155 to reference the actualplaywright-clicommand set (open/goto/screenshot) instead ofbrowser_navigate/browser_take_screenshotgh aw compile/make recompile) so the.lock.ymlreflects the corrected promptSuggested Agent
Developer (workflow markdown change + recompile)
Estimated Effort
Quick (< 1 hour)
Data Source
DeepReport Intelligence Briefing, 2026-10-01 ~06:44Z cycle, sourced from Documentation Noob Test Report discussion #64676 (2026-10-01).