watching-videos
Watch Skill processes any video—URLs, local files, streams, recordings—by extracting frames, running OCR, and transcribing audio into a unified searchable index. Query the index directly instead of re-watching, and batch-process playlists or folders in one pass.
Watch Skill extracts frames, transcripts, and OCR text from videos into a searchable index for instant content analysis.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-07-12
Watch Skill extracts frames, transcripts, and OCR text from videos into a searchable index for instant content analysis. Watch Skill processes any video—URLs, local files, streams, recordings—by extracting frames, running OCR, and transcribing audio into a unified searchable index. Query the index directly instead of re-watching, and batch-process playlists or folders in one pass.
Use it when
- Yes.
- watching-videos processes multiple videos or playlists in batch mode.
Verify before relying
Read SKILL.md below before installing (1 file). Open directory: indexed for reading, not audited.
Install
oxbshw/watch-skill/watching-videos · repository language: Python
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What can watching-videos extract and index from video content?
watching-videos extracts and indexes video content by processing frames, running OCR on text within those frames, and transcribing audio into a unified searchable index. This lets you query the index directly instead of re-watching videos, making it easy to find specific moments, text, or topics across your video library.
Can watching-videos transcribe meeting recordings automatically?
Yes. watching-videos automatically transcribes video recordings, including meeting recordings, by extracting and processing the audio track. The resulting transcript becomes part of the searchable index, so you can find specific discussion points or phrases without manually reviewing the entire recording.
How does watching-videos handle batch processing of videos?
watching-videos processes multiple videos or playlists in batch mode, allowing you to submit a folder of video files or a playlist URL in one pass. The skill indexes all content together, creating a unified searchable database across the entire batch rather than requiring individual file processing.
What video sources does watching-videos support?
watching-videos handles any video format: URLs (including YouTube and TikTok links), local files (MP4, etc.), screen captures, meeting recordings, lecture videos, HLS streams, and playlists. Whether your content is online or stored locally, watching-videos can extract frames, transcribe audio, and index it for search.
Can I search video content by phrase or timestamp?
Yes. watching-videos lets you search and query the indexed video content by phrase or timestamp. Instead of scrubbing through recordings manually, you can search for specific text, topics, or moments across one or multiple videos at once, with results linked back to their source frames and timestamps.
Does watching-videos work offline or require internet?
watching-videos can transcribe video offline locally, processing video files stored on your machine without requiring an internet connection. This makes it ideal for sensitive recordings, large batches, or environments where streaming isn't practical. The skill is released under the MIT license.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
Watching videos
You don't have a video input; this skill gives you one. Anything a user hands you — a YouTube link, a TikTok, a lecture, a meeting recording, a screen capture, an .mp4 on disk — goes through the same pipeline: frames (scene-aware, deduplicated), OCR, transcript (captions first, local Whisper offline fallback), all persisted into one index.
Before watching: check the index
watch-skill list
If the video was already analyzed — this session or any earlier one —
do NOT watch it again. Ask it directly (see the asking-with-evidence
skill):
watch-skill ask <video_id> "<question>"
One video
```bash watch-skill watch "<url-or-path>"
(truncated - see the full file via the links below)
File tree — 1 file
adapters/claude-skill/skills/watching-videos/SKILL.md
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Extract and index video content for searchable analysis”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
Pull structured data from videos you've already watched: titled chapters with timestamps for long content, pinpointed bug reports with error frames and OCR text, or opening-hook analysis scored across attention, pacing, visuals, and on-screen text. All answers come straight from the persistent index with no extra model calls.
Video Memory maintains a searchable index of every video watched on your machine, preserving frames, transcripts, and extracted answers in persistent storage. Search across all indexed videos by keyword or semantic meaning, drill into specific moments, or synthesize answers from insights spanning multiple videos. Batch-import playlists or folders to build your video library quickly.
Watch Skill processes videos locally by extracting scene-aware frames, running OCR on text within them, and transcribing audio via offline Whisper or captions. It builds a searchable index so follow-up questions are answered without re-processing the video. No API keys required for core features; vision and STT are optional and configurable.
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.
When a video answer goes wrong, report the correction and Watch Skill records it as a lesson stored locally. The system classifies the mistake, injects it into similar future questions, and can re-ask the original question to verify the fix worked. Check what's been learned or review token savings anytime.
Turn your video analysis into a self-contained HTML file that anyone can open without needing Watch Skill or internet access. The page bundles frames, transcript, OCR text, and all your cached Q&A with timestamped evidence in one shareable document—ideal for bug tickets, code reviews, or handing off findings to teammates.
More skills asking-with-evidence (MIT) · claude-real-video (MIT) · watch (MIT) · Video Watch (NOASSERTION) · youtube-downloader (MIT) · claude-real-video-for-agents (MIT)