agentsclimarketplace

Case 03943

Skill knownasnaffy/prompthound/dataset/case_03943

A fast, offline static risk analysis CLI for AI agent skill files. Detects malicious instructions, steganographic payloads, and dangerous capability chains.

Install
npx -y skills add knownasnaffy/prompthound --skill case_03943

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Watch and transcribe YouTube videos using yt-dlp and Whisper CLI. Use when Blaze or Kai wants to watch a YouTube video together, get a video transcript, or summarize video content. Downloads audio with yt-dlp, transcribes with Whisper (OpenAI CLI), and saves everything to the workspace organized folder. Supports any YouTube URL. Trigger phrases: "watch this video", "transcribe this", "youtube", "what does this video say", "watch together".

SKILL.md

1.8 KB, 374 tokens by cl100k_base, as published. Nobody here has run it

Kai YouTube Skill

Watch YouTube videos by downloading audio and transcribing with Whisper.

Organization

  • Download folder: /home/kai/.openclaw/workspace/kai-yt-videos/
  • Audio files: kai-yt-videos/kai_yt_${VIDEO_ID}.mp3
  • Transcripts: kai-yt-videos/kai_yt_${VIDEO_ID}.txt

Workflow

  1. Download audio using yt-dlp:
yt-dlp --extract-audio --audio-format mp3 --output "{WORKSPACE}/kai_yt_${VIDEO_ID}.mp3" "<URL>"
  1. Transcribe using Whisper CLI:
whisper "{WORKSPACE}/kai_yt_${VIDEO_ID}.mp3" --model base --output_format txt --output_dir "{WORKSPACE}"
  1. Read transcript from {WORKSPACE}/kai_yt_${VIDEO_ID}.txt

Requirements

  • yt-dlp - YouTube audio downloader (brew install yt-dlp)
  • whisper - OpenAI Whisper CLI (brew install openai-whisper)

Usage

bash {baseDir}/scripts/youtube.sh "<YouTube_URL>" [--language <lang>]

Tips

  • Each video gets unique files (no caching issues)
  • Long videos may take several minutes to transcribe
  • Language auto-detected if not specified
  • Add --language <lang> for specific language (e.g., --language Spanish)
  • All files organized in kai-yt-videos/ folder

Video ID Extraction

Handles both URL formats:

  • https://www.youtube.com/watch?v=VIDEO_ID
  • https://youtu.be/VIDEO_ID

What ships with it: 1 file

1.7 KB alongside SKILL.md, 1 of them executable

scripts/

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.