agentleFS
matheus-rib1108matheus-rib1108/claude-skills-br/skills/video-link/SKILL.md

video-link

Skill0 starsChanged 31 days agoMITView on GitHub

Watches a video from a link or file — Instagram Reels/posts, TikTok, YouTube, Shorts, X, Facebook, Vimeo and 1000+ sites. Downloads it, transcribes the speech (local Whisper, no API key) and extracts keyframes so Claude can see and hear it. Use when the user sends a video link or file and asks what it says, to summarize it, take notes, copy the idea, or "olha esse vídeo / vê esse reels / esse tiktok".

  • Installs packages

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Check tools once: yt-dlp --version && ffmpeg -version && python3 -c "import faster_whisper".

    line 17
  2. Very specific:

    Run the script. Use a new folder per video in the scratchpad (or videos/<name> if the user wants to keep it).

    line 20
  3. Very specific:

    Read summary.json and transcript.txt.

    line 22
  4. Very specific:

    If they use Obsidian, offer to save the notes there (skill obsidian-markdown).

    line 25

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Video link

python3 <this-skill-dir>/scripts/watch.py "<url-or-file>" <out_dir>

Output in <out_dir>: transcript.txt (timestamped speech), frames/ (one frame every 2 s, max 30), summary.json (title, author, caption, duration, views, likes, language).

Steps

  1. Check tools once: yt-dlp --version && ffmpeg -version && python3 -c "import faster_whisper". Missing → pip install -U yt-dlp faster-whisper "av<17" (PyAV 17+ quebra o faster-whisper 1.2) and install ffmpeg (Windows: winget install ffmpeg; macOS: brew install ffmpeg).
  2. Run the script. Use a new folder per video in the scratchpad (or videos/<name> if the user wants to keep it).
  3. Read summary.json and transcript.txt. Look at a handful of frames (start, middle, end, plus any moment the transcript points to) — don't open all 30 unless needed.
  4. Answer what the user asked: summary, step-by-step, quotes with timestamps, ideas to copy. If they use Obsidian, offer to save the notes there (skill obsidian-markdown).

When it fails

  • Instagram/TikTok asks for login or "rate-limit" → retry with the user's browser session: add --cookies-from-browser chrome (or firefox, edge). This reads cookies locally; nothing is sent anywhere else. Only on the user's own computer.
  • Private video → only works if the user's logged-in account can see it.
  • Cloud session with blocked network (HTTP 403 from proxy) → tell the user to run it on their computer or send the video file itself; the script accepts a file path.
  • yt-dlp breaks often when sites change → pip install -U yt-dlp first.

Respect copyright: summarizing and taking notes is fine; don't re-upload other people's videos.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.