{"enrichment":{"faq":[{"a":"Recovering from Errors helps you diagnose and repair Watch Skill failures. Start by running the built-in doctor command, which identifies issues with ffmpeg, yt-dlp, video acquisition, provider authentication, and local models. The doctor will return structured error codes that pinpoint the root cause. Apply the stated fixes, then retry your original operation. If the error persists, report the structured error code without looping or changing security policies.","q":"Watch skill command failed\u2014how to fix it?"},{"a":"Recovering from Errors addresses ffmpeg dependency issues through the doctor command. When ffmpeg is missing, the doctor detects this and provides repair instructions. Install ffmpeg using your system's package manager (apt, brew, or equivalent), verify the installation, then retry your watch command. The doctor command will confirm ffmpeg is properly detected before you proceed with video operations.","q":"How do I troubleshoot ffmpeg missing error recovery?"},{"a":"Recovering from Errors includes yt-dlp troubleshooting as a core repair pathway. Run the doctor command to confirm yt-dlp is missing, then install it via pip or your package manager. After installation, verify the tool is accessible from your command line, then retry your video download or watch operation. The doctor will validate that yt-dlp is properly configured before you continue.","q":"What should I do if yt-dlp is not found?"},{"a":"Recovering from Errors guides you through provider authentication issues. The doctor command detects when API keys are missing, expired, or misconfigured. Check your provider credentials, ensure they are correctly set in your environment or configuration file, and verify the API key has not been revoked. Retry your operation after updating credentials. If errors persist, report the structured error code for further investigation.","q":"How do I resolve provider authentication and API key errors?"},{"a":"Recovering from Errors helps diagnose video acquisition failures. Run the doctor command to identify whether the issue stems from ffmpeg, yt-dlp, provider authentication, corrupt caches, or local model problems. Clear any stale locks or corrupt caches if flagged, repair missing dependencies, and verify provider credentials. Retry your watch operation once fixes are applied. Report any persistent structured errors without changing security policies.","q":"Why can't it watch this video and how do I fix it?"},{"a":"Recovering from Errors describes the doctor command as your primary diagnostic tool. It scans for broken ffmpeg and yt-dlp dependencies, validates provider authentication and API key configuration, detects corrupt caches and stale locks, and checks local vision model and memory health. The doctor returns structured error codes that guide you to specific fixes. Run it whenever a watch command fails to identify the exact issue and next steps.","q":"What does the watch skill doctor command do?"}],"shadow_tags":["error-diagnosis","dependency-repair","troubleshooting-guide","system-health-check","recovery-workflow","infrastructure-debugging","configuration-validation","state-restoration","failure-handling"],"summary_rewrite":"Recovering from Errors guides you through diagnosing Watch Skill failures using the built-in doctor command, which identifies issues with ffmpeg, yt-dlp, video acquisition, provider authentication, and local models. Apply the stated fixes, retry your original operation, and report any persistent structured errors without looping or changing security policies."},"files":[{"bytes":1818,"path":"adapters/claude-skill/skills/recovering-from-errors/SKILL.md","sha256":"38665e6e773479545334931eb86bb9eff773bbc7442d07b2d8c9395517689d02","url":"https://skillfed.io/files/oxbshw/watch-skill/recovering-from-errors/ff8b1b56/SKILL.md"}],"id":"oxbshw/watch-skill/recovering-from-errors","links":{"html":"https://skillfed.io/oxbshw/watch-skill/recovering-from-errors","md":"https://skillfed.io/oxbshw/watch-skill/recovering-from-errors.md","repo":"https://github.com/oxbshw/watch-skill"},"meta":{"agents_supported":[],"first_seen":"2026-07-28","forks":36,"language":"Python","last_updated":"2026-07-12","license":"MIT","name":"recovering-from-errors","publisher":"oxbshw","stars":236},"relations":{"categories":["video-processing","devops","error-recovery"],"similar":[{"id":"oxbshw/watch-skill/configuring-vision"},{"id":"oxbshw/watch-skill/watching-videos"},{"id":"oxbshw/watch-skill/watch"},{"id":"oxbshw/watch-skill/the-loop"},{"id":"oxbshw/watch-skill/learning-from-mistakes"},{"id":"oxbshw/watch-skill/video-memory"},{"id":"oxbshw/watch-skill/extracting-structure"},{"id":"oxbshw/watch-skill/sharing-results"},{"id":"oxbshw/watch-skill/asking-with-evidence"},{"id":"minicoohei/ai-agent-camp/video-analyzer"}]},"slug":{"owner":"oxbshw","repo":"watch-skill","skill":"recovering-from-errors"},"version":"ff8b1b56"}