Skip to content
Back to search

watch

Give Claude a video input. /watch downloads from YouTube/Instagram/X/Vimeo/any yt-dlp site, extracts frames, and transcribes locally with mlx-whisper — no API key. Fork of bradautomates/claude-video.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/mathiaschu--watch/badge.svg)](/listings/mathiaschu--watch)

HTML

<a href="/listings/mathiaschu--watch"><img src="/api/listings/mathiaschu--watch/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

**Claude Code (plugin):**

/plugin marketplace add mathiaschu/claude-video
/plugin install watch@claude-video

**Generic skill folder (Codex, etc.):**

git clone https://github.com/mathiaschu/claude-video.git ~/.claude/skills/watch

Then run once to install dependencies (or let the first `/watch` do it):

python3 ~/.claude/skills/watch/scripts/setup.py

> **Windows:** use `python` instead of `python3` (the `python3` alias is a Microsoft Store stub). `ffmpeg` + `yt-dlp` install via the `winget` commands the setup script prints; for transcription use `pip install openai-whisper` (mlx-whisper is Apple Silicon only).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 21 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "claude-video",
  3. "metadata": {
  4. "description": "Give Claude a video input. /watch downloads from any platform, extracts frames, transcribes locally (no API key), and hands it all to Claude."
  5. },
  6. "owner": {
  7. "name": "Mathias Schusterman"
  8. },
  9. "plugins": [
  10. {
  11. "name": "watch",
  12. "description": "Watch a video (URL or local path) from YouTube, Instagram, X, Vimeo, TikTok or any yt-dlp site. Downloads with yt-dlp, extracts frames with ffmpeg, transcribes via captions or local mlx-whisper.",
  13. "author": {
  14. "name": "Mathias Schusterman"
  15. },
  16. "source": "./",
  17. "category": "productivity",
  18. "homepage": "https://github.com/mathiaschu/claude-video"
  19. }
  20. ]
  21. }
README.md · README · 114 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # /watch — give Claude eyes for video
  2. **Let Claude watch a video from almost anywhere — and transcribe it 100% on your machine, no API key.**
  3. Paste a URL (YouTube, Instagram, X/Twitter, Vimeo, TikTok, Loom, and ~1800 more) or a local file, ask a question, and Claude downloads the video, extracts frames, pulls a timestamped transcript, and *reads every frame as an image*. By the time it answers, it has **seen** the video and **heard** the audio.
  4. ```
  5. /watch https://youtu.be/dQw4w9WgXcQ what happens at the 30 second mark?
  6. ```
  7. > This is a fork of [bradautomates/claude-video](https://github.com/bradautomates/claude-video). The big change: **transcription runs locally** via [mlx-whisper](https://github.com/ml-explore/mlx-examples/tree/main/whisper) (Apple Silicon) or [openai-whisper](https://github.com/openai/whisper) (CPU) instead of a paid Whisper API. No keys, no config file, no audio ever leaves your machine. Also adds browser-cookie support so login-gated sources (Instagram, X) work.
  8. ---
  9. ## Install
  10. **Claude Code (plugin):**
  11. ```
  12. /plugin marketplace add mathiaschu/claude-video
  13. /plugin install watch@claude-video
  14. ```
  15. **Generic skill folder (Codex, etc.):**
  16. ```bash
  17. git clone https://github.com/mathiaschu/claude-video.git ~/.claude/skills/watch
  18. ```
  19. Then run once to install dependencies (or let the first `/watch` do it):
  20. ```bash
  21. python3 ~/.claude/skills/watch/scripts/setup.py
  22. ```
  23. > **Windows:** use `python` instead of `python3` (the `python3` alias is a Microsoft Store stub). `ffmpeg` + `yt-dlp` install via the `winget` commands the setup script prints; for transcription use `pip install openai-whisper` (mlx-whisper is Apple Silicon only).
  24. ## Requirements
  25. - **Python 3.9+**
  26. - **ffmpeg** + **yt-dlp** — auto-installed via Homebrew on macOS; on Linux/Windows the installer prints the exact command.
  27. - **A local Whisper engine** (only needed for videos *without* captions):
  28. - **mlx-whisper** — `pip3 install mlx-whisper` — preferred on Apple Silicon, runs on the GPU/Neural Engine.
  29. - **openai-whisper** — `pip3 install openai-whisper` — CPU fallback, cross-platform.
  30. There is **no API key**. Captions cover most public videos for free; when a video has no captions, the audio is transcribed locally.
  31. ## How it works
  32. 1. **You paste a video and a question.** A URL (anything yt-dlp supports) or a local path (`.mp4`, `.mov`, `.mkv`, `.webm`, …).
  33. 2. **`yt-dlp` downloads it** into a temp working directory. Local files are probed in place, no download.
  34. 3. **`ffmpeg` extracts frames at an auto-scaled rate.** Duration-aware budget — ≤30s gets ~30 frames, 1-3min ~60, 3-10min ~80, longer 100 sparsely. Hard ceilings: 2 fps, 100 frames. JPEGs at 512px wide (bump with `--resolution 1024` to read on-screen text).
  35. 4. **The transcript comes from one of two places.** First: yt-dlp pulls native captions (free, instant). Fallback: a mono 16 kHz audio clip is transcribed **locally** with mlx-whisper / openai-whisper.
  36. 5. **Frames + transcript are handed to Claude.** It `Read`s each frame as an image and aligns them to the timestamped transcript.
  37. 6. **Claude answers grounded in what's on screen and in the audio** — not the title, not a guess.
  38. 7. **Cleanup.** The script prints its working directory; Claude removes it when you're done.
  39. ## Login-gated sources (Instagram, X, private videos)
  40. Public videos download with no auth. For sources that require a login — Instagram, X/Twitter, age-restricted or private/unlisted YouTube — `/watch` needs **your own** browser cookies so yt-dlp can authenticate as you. You don't have to set anything up in advance: just run `/watch <url>` normally, and if it fails with a login/private error, do this.
  41. ### The easy way — borrow cookies from your browser
  42. 1. Make sure you're **logged into the site** (e.g. Instagram) in a normal browser.
  43. 2. Tell Claude which browser it is, or run it directly:
  44. ```
  45. /watch https://www.instagram.com/reel/XXXX/ --cookies-from-browser chrome
  46. ```
  47. Supported: `chrome`, `firefox`, `safari`, `edge`, `brave`, `chromium`, `opera`, `vivaldi`.
  48. **macOS gotchas (the usual culprits):**
  49. - **Chrome must be fully quit** (Cmd-Q, not just the window) — it locks its cookie database while running.
  50. - A **Keychain prompt** may appear ("… wants to use confidential information stored in Chrome Safe Storage"). Click **Always Allow** — that's macOS letting the tool decrypt Chrome's cookies on your own machine.
  51. - **Safari** requires giving your terminal / Claude Code **Full Disk Access** (System Settings → Privacy & Security → Full Disk Access).
  52. - When in doubt, **Firefox** tends to "just work" without closing it.
  53. ### The reliable way — export a cookies.txt
  54. If browser extraction keeps fighting you, export the cookies manually (works everywhere):
  55. 1. Install a cookies exporter: **"Get cookies.txt LOCALLY"** (Chrome/Edge) or **"cookies.txt"** (Firefox) — both open-source.
  56. 2. Open and log into the site (e.g. `instagram.com`).
  57. 3. Click the extension → **Export** → save it, e.g. `~/Downloads/cookies.txt`.
  58. 4. Run:
  59. ```
  60. /watch https://www.instagram.com/reel/XXXX/ --cookies ~/Downloads/cookies.txt
  61. ```
  62. ### Privacy
  63. **Your cookies are read live from your own machine and piped straight into yt-dlp.** The skill never copies, stores, logs, or transmits them. If you exported a `cookies.txt`, it stays on your disk — delete it whenever you want.
  64. ## Useful flags
  65. | Flag | What it does |
  66. |------|--------------|
  67. | `--start T` / `--end T` | Focus on a section (`SS`, `MM:SS`, or `HH:MM:SS`); denser frame rate |
  68. | `--max-frames N` | Lower the frame cap for a tighter token budget |
  69. | `--resolution W` | Frame width in px (default 512; 1024 to read on-screen text) |
  70. | `--fps F` | Override auto-fps (clamped to 2 fps) |
  71. | `--cookies-from-browser B` | Read cookies from a local browser for login-gated sources |
  72. | `--cookies FILE` | Use a Netscape-format `cookies.txt` instead |
  73. | `--whisper mlx\|openai-whisper` | Force a specific local engine |
  74. | `--no-whisper` | Skip transcription entirely (frames only) |
  75. ## Privacy
  76. - **No API key, no account, no telemetry.** There is no `.env`, no config file, no secret of any kind.
  77. - **Audio never leaves your machine** — transcription is fully on-device.
  78. - The only outbound network traffic is yt-dlp fetching the video/captions from the source URL, and a one-time ~1.5 GB Whisper model download from Hugging Face (cached afterwards).
  79. ## Credits
  80. Original skill: [bradautomates/claude-video](https://github.com/bradautomates/claude-video) (MIT). This fork swaps the Whisper API for local transcription and adds browser-cookie support. Licensed MIT — see [LICENSE](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.