Video Debug
Use when the user references a video, screencast, screen recording, or any path ending in .mp4, .mov, .webm, .mkv, or .gif — especially when asking to debug a UI bug, visual glitch, or reproduction recording. Extracts key frames via ffmpeg scene detection so the agent can visually inspect what went wrong.
Install
npx skills add vladzima/video-debug --skill video-debuggit clone --depth 1 https://github.com/vladzima/video-debug.git /tmp/video-debug
mkdir -p ~/.claude/skills
cp -r /tmp/video-debug/. ~/.claude/skills/video-debugPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Video Debug Protocol
When the user provides a path to a video file (`.mp4`, `.mov`, `.webm`, `.mkv`, or `.gif`), follow this protocol.
## Step 1: Extract frames
Run the extractor against the video path the user gave you:
```bash
bash scripts/extract.sh <video-path>
```
The script will:
1. Verify `ffmpeg` is installed, and if not, offer to install it (you may need to relay a `[y/N]` prompt to the user).
2. Probe the video's duration, resolution, and estimated scene-change count.
3. Decide whether the video is "small enough to process directly" or "large enough to need a strategy."
## Step 2: Handle large videos
If `extract.sh` exits with code **10**, the video is large (currently: duration > 60 seconds). The script will have printed a single JSON line to stdout with this shape:
```json
{"large": true, "duration_seconds": 248.5, "resolution": "1920x1080"}
```
Surface these numbers to the user and present them with these four options:
- **a)** Process all frames as-is (may use many tokens).
- **b)** Sample down to ~30 evenly-distributed frames.
- **c)** Focus on a specific time range — ask the user for `start–end` in `MM:SS` or seconds.
- **d)** Use a stricter scene threshold (fewer, more distinct frames).
Then re-run with the chosen strategy:
```bash
bash scripts/extract.sh <video-path> --strategy=<a|b|c|d> [--range=START-END]
```
For option **c**, pass `--range=MM:SS-MM:SS` (or `--range=12-30` for raw seconds).
## Step 3: Read the timelineDetails
- Skill name
video-debug- Repository
- vladzima/video-debug
- License
- MIT
- Repo updated
- 19 May 2026
FAQ
What does the Video Debug skill do?
Use when the user references a video, screencast, screen recording, or any path ending in .mp4, .mov, .webm, .mkv, or .gif — especially when asking to debug a UI bug, visual glitch, or reproduction recording. Extracts key frames via ffmpeg scene detection so the agent can visually inspect what went wrong.
How do I install the Video Debug skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add vladzima/video-debug --skill video-debug. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Video Debug popular?
It lives in vladzima/video-debug, which has 5 GitHub stars and 0 forks.
Related skills
Video GIF Studio
5miles990/video-gif-studio
Plan scripts and storyboards, analyze music cue points, create character animation and game sprite exports, or edit existing video. Use for GIF, APNG, transparent frames, sprite sheets and video deliverables from text, images, video or music. Grok is the default motion source; other available video tools, image keyframes and local media are alternatives.
Tiktok Ugc Seedance
5xigua0626/tiktok-ugc-seedance
Convert a target ecommerce product and one TikTok competitor video into a TikTok Shop UGC strategy package: product positioning, category/search keywords for finding competitor videos, competitor video transcript and conversion-structure breakdown, 15-second English UGC voiceover script, visual shot plan, and Seedance 2.0 prompt. Use when the user provides a product link/image plus a TikTok reference link, asks to analyze a competitor TikTok video, write UGC TikTok scripts, or turn the script into Seedance prompts for cross-border ecommerce ads.
Video Reverse Engineering
5whaojie797-design/video-reverse-engineering
逆向拆解一条视频的拍摄与制作方法：给定一个视频链接，下载并抽取关键帧、字幕与音频，产出(1)分镜脚本、(2)可直接用于AI绘图/生图工具的制图提示词（含图生视频运镜提示）、(3)完整复刻/仿拍制作指南。当用户丢来一个视频网址（YouTube、B站/Bilibili、抖音、快手、小红书、TikTok、Instagram Reels等）并表达"分析这个视频""这个视频怎么拍/怎么做的""拆解一下这个视频""复刻/仿拍这个视频""学习这个视频的运镜和风格"等意图，或英文 analyze/breakdown/reverse-engineer/recreate this video 之类请求时，都要主动使用本Skill——即便用户只发了一个链接、没有说"拆解"两个字，只要意图是理解或复刻视频的制作方法而不是单纯想要内容摘要，就应触发。
Superpartner
5coffeeandburger72/superpartner
Build a living profile of your partner from chat history. Analyze personality, interests, important dates, and generate gift ideas and dating plans. Supports WhatsApp, WeChat, and LINE exports.
Xiaomimimo Extended
5hecaodong/xiaomimimo-extended
Xiaomi MiMo API toolkit: web search, image / audio / video understanding, and TTS speech synthesis. 调用小米 MiMo API 的隐藏能力 —— 联网搜索、图片识别、音频转写、视频分析、语音合成(TTS)。 Trigger phrases — 触发词: EN: mimo search, web search, image analysis, screenshot OCR, audio transcription, meeting summary, video analysis, tutorial extract, text to speech, voice clone 中: mimo 搜一下 / 联网搜 / 看看这张图 / 这段录音说了啥 / 转写 / 提取视频步骤 / 合成语音 / 念出来 / 克隆音色
Sansheng Gemini Video
5sanshengai/sansheng-gemini-video
Use when 用户要用 Gemini 原生多模态联合分析本地或在线视频的画面、声音与时序，支持 YouTube 直读，以及按需下载 B站、小红书、抖音或经一次性授权解析微信视频号；触发词：分析视频、录屏质检、视频链接分析、批量验收。只要字幕、只下载、或要剪辑出片（交给 sandy-video）时不用此 Skill。
Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.