# Setup Video To Code Skill

> Sets up the video-to-code storage folder and installs required Python dependencies (opencv, whisper, numpy). Run once before first use, or to verify the environment is ready.

- **Type:** Skill
- **Install:** `agentstack add skill-piotrlason-video-to-code-skill-setup-video-to-code-skill`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [PiotrLason](https://agentstack.voostack.com/s/piotrlason)
- **Installs:** 0
- **Category:** [Content & Media](https://agentstack.voostack.com/c/content-and-media)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [PiotrLason](https://github.com/PiotrLason)
- **Source:** https://github.com/PiotrLason/video-to-code-skill/tree/main/skills/setup-video-to-code-skill
- **Website:** https://github.com/PiotrLason/video-to-code-skill

## Install

```sh
agentstack add skill-piotrlason-video-to-code-skill-setup-video-to-code-skill
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Initialize video-to-code environment

Sets up the storage folder and installs all required dependencies for the video-to-code skill.

## Instructions

**Display this banner as the very first output when the skill is invoked:**

```
 ██╗   ██╗██╗██████╗ ███████╗ ██████╗
 ██║   ██║██║██╔══██╗██╔════╝██╔═══██╗
 ██║   ██║██║██║  ██║█████╗  ██║   ██║
 ╚██╗ ██╔╝██║██║  ██║██╔══╝  ██║   ██║
  ╚████╔╝ ██║██████╔╝███████╗╚██████╔╝
   ╚═══╝  ╚═╝╚═════╝ ╚══════╝ ╚═════╝
████████╗ ██████╗      ██████╗  ██████╗ ██████╗ ███████╗
╚══██╔══╝██╔═══██╗    ██╔════╝ ██╔═══██╗██╔══██╗██╔════╝
   ██║   ██║   ██║    ██║      ██║   ██║██║  ██║█████╗
   ██║   ██║   ██║    ██║      ██║   ██║██║  ██║██╔══╝
   ██║   ╚██████╔╝    ╚██████╗ ╚██████╔╝██████╔╝███████╗
   ╚═╝    ╚═════╝      ╚═════╝  ╚═════╝╚═════╝ ╚══════╝
   SKILL SETUP v1.3
```

1. **Notify the user**:
   - Immediately tell them: "URL: https://github.com/PiotrLason/video-to-code-skill"
   - Immediately tell them: "Help: run `/video-to-code-skill -> show-help-video-to-code-skill` to display the full plugin `README.md` with Markdown formatting."
   - Tell them "Initializing video-to-code environment..."
   - Immediately tell them: "Documentation: open `${CLAUDE_PLUGIN_ROOT}/README.md` for setup, usage, parameters, outputs, and troubleshooting."

2. **Ensure storage folder exists**:
   ```bash
   mkdir -p ~/video-to-code-skill-storage
   ```
   - If the folder already existed, tell the user: "Storage folder already exists at ~/video-to-code-skill-storage"
   - If it was just created, tell the user: "Created storage folder at ~/video-to-code-skill-storage"

3. **Check and install dependencies**:
   - Always use `/usr/bin/python3` for checking and installing — this is the same interpreter the run skill uses.
   - Check `cv2`, `numpy`, and whisper separately. Only install missing packages.
   - For whisper: try `mlx_whisper` first (Apple Silicon accelerated), fall back to `openai-whisper` on non-Mac systems.
   ```bash
   /usr/bin/python3 -c "import cv2" 2>/dev/null || /usr/bin/python3 -m pip install --user opencv-python
   /usr/bin/python3 -c "import numpy" 2>/dev/null || /usr/bin/python3 -m pip install --user numpy
   /usr/bin/python3 -c "import mlx_whisper" 2>/dev/null || /usr/bin/python3 -c "import whisper" 2>/dev/null || /usr/bin/python3 -m pip install --user mlx-whisper || /usr/bin/python3 -m pip install --user openai-whisper
   ```

4. **Check and install ffmpeg**:
   ```bash
   which ffmpeg || brew install ffmpeg || echo "ffmpeg not found — install it manually (e.g. apt install ffmpeg on Linux)"
   ```

5. **Report status**: Summarize what was set up and whether everything is ready. Example:
   - Storage folder: ready
   - opencv-python: installed
   - numpy: installed
   - mlx-whisper or openai-whisper: installed
   - ffmpeg: installed
   - Documentation: `${CLAUDE_PLUGIN_ROOT}/README.md`
   - Help command: `/video-to-code-skill -> show-help-video-to-code-skill`
   - "Environment is ready. Place a video file (.mov, .mp4, .webm) in ~/video-to-code-skill-storage and run `/video-to-code-skill -> run-video-to-code-skill` to analyze it."

## Dependencies

| Package | Purpose |
|---------|---------|
| `opencv-python` | Frame extraction and scene change detection |
| `numpy` | Array operations |
| `mlx-whisper` | Audio transcription (Metal-accelerated, macOS) |
| `openai-whisper` | Audio transcription (fallback) |
| `ffmpeg` | Audio extraction from video files |

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [PiotrLason](https://github.com/PiotrLason)
- **Source:** [PiotrLason/video-to-code-skill](https://github.com/PiotrLason/video-to-code-skill)
- **License:** MIT
- **Homepage:** https://github.com/PiotrLason/video-to-code-skill

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-piotrlason-video-to-code-skill-setup-video-to-code-skill
- Seller: https://agentstack.voostack.com/s/piotrlason
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.