{"slug":"ui-review","title":"ui-review","tags":["site-building","frontend","review"],"agent_summary":"Workflow 1","trigger_phrases":["ui-review","review guide","run qa","validate stories","ui review","parallel qa"],"runnable":false,"markdown":"**Trigger:** Use when \"run QA\", \"validate stories\", \"ui review\", \"parallel QA\". Multi-page story validation with parallel browser agents and pass/fail report.\n\n### Workflow\n\n1. DISCOVER — Glob for all story files in STORIES_DIR. Parse each story file to extract name, URL, and ordered workflow steps.\n\n2. SETUP — Create the output directory structure: OUTPUT_DIR/ with per-story subdirs and summary-report.md.\n\n3. SPAWN — For each discovered story, spawn a browser-qa-agent using the Task tool in parallel. The meta-prompt to each sub-agent includes: target URL, output directory, step-by-step instructions to perform the action, take a screenshot, and record PASS or FAIL. Run all story agents in parallel — launch them all in a single message with multiple Task tool calls.\n\n4. COLLECT — After all agents complete, collect their results. Each agent returns overall PASS/FAIL status, step-by-step results, and screenshot file paths.\n\n5. REPORT — Compile all results into a unified report with: date, stories tested, passed count, failed count, per-story results table, failed story details with screenshots.\n\n6. CLEANUP — Confirm all browser sessions are closed. Report the final summary to the user. If any stories failed, suggest next steps.\n\n### Key Principles\n\n- Parallel execution — All stories run simultaneously via separate agents\n- Meta-prompt engineering — You teach each sub-agent HOW to validate by embedding the story details into a structured prompt\n- Fail loudly — Every failure gets detailed reporting with screenshots\n- Hands-off — Once spawned, agents work autonomously and report back via task results\n\n---\n","html":"<p><strong>Trigger:</strong> Use when \"run QA\", \"validate stories\", \"ui review\", \"parallel QA\". Multi-page story validation with parallel browser agents and pass/fail report.</p>\n<h3>Workflow</h3>\n<ol>\n<li>\n<p>DISCOVER — Glob for all story files in STORIES_DIR. Parse each story file to extract name, URL, and ordered workflow steps.</p>\n</li>\n<li>\n<p>SETUP — Create the output directory structure: OUTPUT_DIR/ with per-story subdirs and summary-report.md.</p>\n</li>\n<li>\n<p>SPAWN — For each discovered story, spawn a browser-qa-agent using the Task tool in parallel. The meta-prompt to each sub-agent includes: target URL, output directory, step-by-step instructions to perform the action, take a screenshot, and record PASS or FAIL. Run all story agents in parallel — launch them all in a single message with multiple Task tool calls.</p>\n</li>\n<li>\n<p>COLLECT — After all agents complete, collect their results. Each agent returns overall PASS/FAIL status, step-by-step results, and screenshot file paths.</p>\n</li>\n<li>\n<p>REPORT — Compile all results into a unified report with: date, stories tested, passed count, failed count, per-story results table, failed story details with screenshots.</p>\n</li>\n<li>\n<p>CLEANUP — Confirm all browser sessions are closed. Report the final summary to the user. If any stories failed, suggest next steps.</p>\n</li>\n</ol>\n<h3>Key Principles</h3>\n<ul>\n<li>Parallel execution — All stories run simultaneously via separate agents</li>\n<li>Meta-prompt engineering — You teach each sub-agent HOW to validate by embedding the story details into a structured prompt</li>\n<li>Fail loudly — Every failure gets detailed reporting with screenshots</li>\n<li>Hands-off — Once spawned, agents work autonomously and report back via task results</li>\n</ul>\n<hr>\n"}