video_toolkit

1# Video Toolkit
2 
3Turn a brief into a finished MP4 with [claude-code-video-toolkit](https://github.com/digitalsamba/claude-code-video-toolkit).
4This skill is written for **any agent runtime**: Claude Code, OpenClaw, Codex, Hermes, a bare API loop with a
5`run_tool` call, or an MCP client. Nothing here depends on how your runtime backgrounds work, because the long
6work runs in a detached job you poll.
7 
8## Who this is for
9 
10You are an agent: someone asks you for a video through a channel (a chat message, a Discord or Slack thread, an
11MCP client) and you make it without a person at your terminal. You report back in that channel. The person asking
12sees your messages and the finished video, never the toolkit.
13 
14## What the toolkit can make: pick a route first
15 
16Every route has one of two support levels. `catalog` gives it per template as `agent_support`.
17 
18- **`job`: one prompt.** `video_job.py` runs it unattended from one spec. You write the scenes; it voices,
19  illustrates, captions, builds and QCs them; you post the MP4. Any runtime can do this, through the CLI or MCP.
20- **`session`: a working session.** Copy a template or example into `projects/<name>/`, edit its code with the
21  skill it names, render with Node, look at the result, fix it, render again. That needs a Claude Code (or
22  Codex-style) session in the toolkit, which is what the toolkit is built around. The session can be a person's, or
23…