Video Clip Extractor
Extract highlight clips from long videos with transcripts, titles, and covers
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 248345b
Located SKILL.md at .claude/skills/video-clip-extractor/SKILL.md (mirrored under .cursor and .trae). Frontmatter parses with name+description and a scoped allowed-tools list. Spot-checked referenced targets via raw fetch: video_orchestrator.py (200), pyproject.toml (200), README.md (200); the core/ modules it imports are all present in the repo tree. Verified the SKILL.md CLI reference against the actual argparse in video_orchestrator.py — every documented flag exists (code even has a few undocumented extras). No security smells: no curl|sh, no base64 blobs, no secret exfiltration; install uses ~/.local/share/openclip and $OLDPWD with no hardcoded machine paths (yt-dlp is pulled from git master, a standard-but-notable supply-chain choice). Could NOT measure output vs baseline: producing highlight clips requires uv sync, ffmpeg, a downloadable video, Whisper transcription, and a live LLM API key — none executable in this context, so no artifact was produced.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 5/10
- Docs & honesty 5/5
What Video Clip Extractor does
Runs an orchestrator pipeline that downloads a Bilibili/YouTube URL or local video, transcribes it via platform subtitles or Whisper, has an LLM score segments for engaging moments, and cuts highlight clips with optional artistic titles, burned subtitles, and cover images. Triggers when the user wants to extract or clip the best moments from long videos, livestreams, or recordings. Requires uv, ffmpeg, and an LLM API key (Qwen/OpenRouter/GLM/MiniMax).
How to install Video Clip Extractor
git clone --depth 1 https://github.com/linzzzzzz/openclip.git /tmp/video-clip-extractor-src
mkdir -p ~/.claude/skills
cp -R /tmp/video-clip-extractor-src/.claude/skills/video-clip-extractor ~/.claude/skills/video-clip-extractor
# Runtime deps (the skill body handles this on first use):
# git clone https://github.com/linzzzzzz/openclip.git ~/.local/share/openclip
# cd ~/.local/share/openclip && uv sync
# Prerequisites: uv (brew install uv / pip install uv), ffmpeg (brew install ffmpeg),
# and ONE LLM API key env var: QWEN_API_KEY | OPENROUTER_API_KEY | GLM_API_KEY | MINIMAX_API_KEY
# Optional: --burn-subtitles needs ffmpeg+libass; --speaker-references needs `uv sync --extra speakers` + HUGGINGFACE_TOKEN
# Run: cd ~/.local/share/openclip && uv run python video_orchestrator.py <video-url-or-path>
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Video Clip Extractor
-
/video-clip-extractorExtract highlight clips from long videos with transcripts, titles, and covers
It also activates on plain-language prompts like these:
-
Extract highlight clips from this livestream -
Cut the best moments from this YouTube video -
Generate a transcript and clip covers for this recording
Frequently asked questions
- Is the Video Clip Extractor skill free?
- Yes. The skill itself is free from linzzzzzz/openclip. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Video Clip Extractor work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Located SKILL.md at .claude/skills/video-clip-extractor/SKILL.md (mirrored under .cursor and .trae). Frontmatter parses with name+description and a scoped allowed-tools list. Spot-checked referenced targets via raw fetch: video_orchestrator.py (200), pyproject.toml (200), README.md (200); the core/ modules it imports are all present in the repo tree. Verified the SKILL.md CLI reference against the actual argparse in video_orchestrator.py — every documented flag exists (code even has a few undocumented extras). No security smells: no curl|sh, no base64 blobs, no secret exfiltration; install uses ~/.local/share/openclip and $OLDPWD with no hardcoded machine paths (yt-dlp is pulled from git master, a standard-but-notable supply-chain choice). Could NOT measure output vs baseline: producing highlight clips requires uv sync, ffmpeg, a downloadable video, Whisper transcription, and a live LLM API key — none executable in this context, so no artifact was produced.
- What is the Video Clip Extractor SkillProof Score?
- 8.0/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 5/5.
- How do I install Video Clip Extractor?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Video Clip Extractor with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.