1.1k
GitHub Stars
4
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 clip-it- _meta.json277 B
- clip.py19.5 KB
- requirements.txt48 B
- SKILL.md4.2 KB
Overview
This skill is the master tool for advanced audio and video processing. It clips and trims media, finds semantically relevant segments, isolates vocals or noise, and can dub content from YouTube URLs or local files. It handles transcription and audio conversion internally for a seamless workflow.
How this skill works
The tool runs a local executable that transcribes audio, performs semantic search over the transcript, and applies audio manipulations (trimming, isolation, dubbing). You provide an input (YouTube link or local path) and a natural-language query describing the segment to extract; optional flags control isolation or dubbing. The process reports an OUTPUT_FILE path on success, which is then retrieved and returned.
When to use it
- Trim or cut a long video into focused clips by topic or time range
- Find a specific spoken segment in a meeting, podcast, or lecture
- Remove background noise or isolate vocals from a recording
- Dub or translate spoken audio into another language
- Create short highlights or summaries as standalone audio clips
- Process YouTube links or local MP3/WAV/MP4 files directly
Best practices
- Provide a clear, concise query describing the segment (e.g., "discussion about budget").
- Give a direct input: a full YouTube URL or an exact local file path.
- When dubbing, supply the two-letter ISO language code (en, hi, es, fr, de, ja, etc.).
- Specify isolation needs explicitly ("isolate vocals" or "remove background noise").
- If no clip is specified, use a generic query like "full audio" to process the entire file.
- Expect and check the script output for a line like: OUTPUT_FILE: /path/to/result.wav
Example use cases
- Find and export the portion of a YouTube talk where they discuss budgeting.
- Clean a noisy interview recording and extract only the interviewee's voice.
- Dub a tutorial video into Spanish for non-English audiences.
- Trim a podcast to a five-minute highlight reel of the main talking points.
- Extract a segment where an exact phrase or sensitive detail is spoken for compliance or review.
FAQ
You can pass YouTube URLs and local MP3, WAV, or MP4 file paths directly.
How do I request dubbing into a specific language?
Include the dub flag and a two-letter ISO code (e.g., --dub "es" for Spanish) or ask to dub into a named language; specify the language code if possible.
How will I receive the processed clip?
On success the tool prints an OUTPUT_FILE path (e.g., /path/to/result.wav). The file at that path is uploaded and returned.
What if the tool fails?
If the executable errors, check the last three lines of its log for diagnostic text and return those lines for troubleshooting.