2.5k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill video-transcript- _meta.json467 B
- SKILL.md4.1 KB
Overview
This skill extracts full transcripts from video content for analysis, summarization, note-taking, or research. It uses a video transcription API to return readable text or structured JSON with timestamps and metadata. The skill supports YouTube links, short IDs, and Shorts, and is optimized for workflows that need searchable text or precise timecodes. Setup guides API key creation and safe local storage for the key.
How this skill works
Provide a video URL or an 11-character YouTube ID and the skill calls the transcript API to retrieve the transcript. You can request plain text with timestamps or structured JSON that includes per-segment start times and durations. The API can also return metadata like title, author, and thumbnail when requested. Common errors are surfaced (bad key, no credits, no transcript, timeout) with recommended next steps.
When to use it
- Transcribe a lecture, interview, or conference talk for study or archiving.
- Create searchable text for video libraries, knowledge bases, or CMS imports.
- Prepare summaries, highlights, or meeting notes from recorded sessions.
- Extract quotes with precise timestamps for citations or video editing.
- Convert video content to accessible written alternatives for users.
Best practices
- Request format=text for readable outputs and format=json when you need exact timestamps.
- Set send_metadata=true to capture title and channel context for citations or metadata fields.
- Summarize long transcripts into key points first; offer full text on demand to save attention and bandwidth.
- Handle common API errors: re-authenticate on 401, top up credits on 402, retry once after 2s on 408.
- Store the API key in a secure local config and back up existing config before writing changes.
Example use cases
- Researcher extracts full transcripts from YouTube talks to run topic modeling and literature review.
- Product manager generates meeting notes and timestamps for stakeholder action items.
- Content creator pulls captions and timestamps to create social media clips and quoted posts.
- Accessibility engineer provides written transcripts for videos lacking captions.
- Archivist archives video metadata and text for a searchable historical collection.
FAQ
You can use full YouTube URLs, youtu.be short links, shorts links, or the bare 11-character video ID.
Which format should I choose: text or json?
Choose text for human-readable transcripts and json when you need per-segment timestamps and precise offsets.