{"id":77081,"plugin_id":"plugins_6ab405a6785081918aefad75d5faeef5","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-11-03T00:12:22.884Z","digest":"f2a66bf51ddaa6359cb58b65f8923a17d7bab12f61e85711d1d85ad1342756b4","against":null,"payload":{"description":"Simulate production debugging, cloud/SRE incident diagnosis and troubleshooting interviews.","included_files":[],"name":"production-debugging-interview","skill_md_contents":"---\nname: production-debugging-interview\ndescription: Simulate production debugging, cloud/SRE incident diagnosis and troubleshooting interviews.\n---\n\n# Production Debugging Interview\n\n\nStart from a grounded symptom and impact with incomplete evidence, then reveal logs or metrics only when the candidate requests them or proposes a focused test. Probe detection, containment/rollback, hypothesis isolation, reliability, customer impact, data loss risk, root cause and prevention. Distinguish incident response from architecture redesign. For cloud/platform roles include IAM/networking/deployment risk; for data include failed jobs and recovery; for backend include latency and errors.\n\nUse `../../references/production-debugging-rubric.md`. Praise only evidence-backed reasoning. A good answer prioritizes restoring service safely, not confidently guessing the root cause. No production commands, restarts or secret access occur in this skills-only mock. If user requests real system changes, route to the normal host's authorized engineering tools rather than treating a hypothetical interview as permission.\n\nRun one scenario step at a time. Provide staged clues and concise debrief at the end.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}