Watch
Watch a video from YouTube, Instagram, X/Twitter, Vimeo, TikTok or any of ~1800 yt-dlp sites (or a local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or local mlx-whisper fallback, no API key), and hands the result to Claude so it can answer questions about what's in the video.
Install
npx skills add mathiaschu/watch --skill watchgit clone --depth 1 https://github.com/mathiaschu/watch.git /tmp/watch
mkdir -p ~/.claude/skills
cp -r /tmp/watch/. ~/.claude/skills/watchPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# /watch — Claude watches a video
You don't have a video input; this skill gives you one. A Python script downloads the video, extracts frames as JPEGs, gets a timestamped transcript (native captions first, then **local mlx-whisper** as fallback — runs on-device, no API and no key), and prints frame paths. You then `Read` each frame path to see the images and combine them with the transcript to answer the user.
## Step 0 — Setup preflight (runs every `/watch` invocation, silent on success)
**Python interpreter:** every `python3 ...` command in this skill is for macOS/Linux. On **Windows**, substitute `python` — the `python3` command on Windows is the Microsoft Store stub and will not run the script.
Before every `/watch` run, verify that dependencies are in place:
```bash
python3 "${CLAUDE_SKILL_DIR}/scripts/setup.py" --check
```
This is a <100ms lookup. On exit 0, the script emits **nothing** — proceed to Step 1 without comment. **Do NOT announce "setup is complete" to the user** — they don't need a status message on every turn. The only acceptable user-visible output from Step 0 is when remediation is required.
On non-zero exit, follow the table:
| Exit | Meaning | Action |
|------|---------|--------|
| `2` | Missing binaries (`ffmpeg` / `ffprobe` / `yt-dlp`) | Run installer |
| `3` | No local whisper engine (`mlx-whisper` / `openai-whisper`) | Run installer, then tell user the `pip3` command it prints |
| `4` | Both missing | Run installer |Details
- Skill name
watch- Repository
- mathiaschu/watch
- License
- MIT
- Repo updated
- 29 May 2026
FAQ
What does the Watch skill do?
Watch a video from YouTube, Instagram, X/Twitter, Vimeo, TikTok or any of ~1800 yt-dlp sites (or a local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or local mlx-whisper fallback, no API key), and hands the result to Claude so it can answer questions about what's in the video.
How do I install the Watch skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add mathiaschu/watch --skill watch. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Watch popular?
It lives in mathiaschu/watch, which has 144 GitHub stars and 20 forks.
Related skills
Hand Drawn Diagrams
85muthuishere/hand-drawn-diagrams
Create hand-drawn Excalidraw diagrams, flows, explainers, wireframes, and page mockups. Default to monochrome sketch output; allow restrained color only for page mockups when the user explicitly wants webpage-like fidelity. Use when the user asks for a diagram, flowchart, wireframe, sketch, visual explanation, mind map, architecture overview, or any Excalidraw-based drawing.
Social Captions
145rediumvex/social-media-caption-generator-claude
Write algorithm-optimized captions/descriptions for social media posts and reels. Supports Instagram (Reels, Feed, Carousel), TikTok, Threads, Facebook, and YouTube Shorts. The user sends material (text, image, script, competitor post, screenshot) and gets ready-to-use captions for ALL platforms at once, in English. Use when asked to "write a caption", "write a description", "caption for instagram", "описание к посту", "подпись для рилса", "описание для тикток", "caption for reels", "social media post", or any caption/description writing task. Also triggers when user sends an image, screenshot
Placecall
110voygr-tech/placecall
PlaceCall lets your agent call US businesses for reservations, inquiries and quotes, and gets through IVRs, holds and transfers. You get a verified outcome and the full transcript. First 250 calls free; you pay only for outcomes.
Find Broll
141louisedesadeleer/b-roll-finder
Source b-roll for a video edit — classify each moment, scope the search, return vetted candidates, place on the word. Use when the user says "find b-roll", "/find-broll", "source clips for this edit", or wants footage/memes/screenshots to lay over a talking-head video.
Qiaomu Codex Imagegen
136joeseesun/qiaomu-codex-imagegen
Generate or edit images with Codex's built-in image generation from any agent, through an MCP tool or a plain CLI. For every image request it first proposes at least four divergent style directions (24 distilled templates, 48 presets, 20 Mondo poster artists), lets the user pick, expands the prompt, generates, and checks the result. Scenario craft for 视频封面, 视频海报, 小红书配图, 公众号封面, X 封面, 朋友圈海报, 书籍封面, 专辑封面, 产品海报, 人物写真, 节气海报, 展览海报, PPT 页面, 图生图. Use for 生图, 配图, 海报, 封面, "给我几个风格", "用 Codex 画一张". Excludes: editing photos with local tools, charts or diagrams from data, UI mockups, video generation, or any
HTML to Video
166wx4766/html-to-video
把 HTML/Canvas 动画逐帧渲染成 MP4（可配乐、可出 GIF、可出透明层）。当用户说「用代码做视频」「HTML 转视频」「导出 MP4/GIF」「做个动态图解/宣传片/数据动画」「逐帧渲染」时使用。自带渲染器与 11 个开箱即用模板（含多镜头讲解片、竖屏 9:16、前后对比、流程图、字幕卡）、图标/人物/运镜素材库，支持透明输出、外部字幕时间轴与 JSON 镜头表，装好依赖后一条命令出片。
Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.