video-production-agentlisted
Install: claude install-skill zobaerio/zobvia
> **Bundled with Zobvia** (extras/video-production-agent/) — see the sub-skill parts of Zobvia's SKILL.md. Source: [kajisho5/video-production-agent](https://github.com/kajisho5/video-production-agent) v0.1.0 (main @ 7162c69), MIT License (LICENSE in this folder). This is the ecosystem's BRAIN: it plans a production, asks the chat model (you) for the reasoning steps, and drives the execution skills Zobvia bundles (ffmpeg-skill, audio-production, color-grading, thumbnail) with validation, approval gates and a recorded provenance trail. Bundled: SKILL.md, CLAUDE.md, `src/`, `schemas/`, `profiles/`, `tools/`, `pyproject.toml`. Not bundled: docs, tests, evals, brand. Runs via `PYTHONPATH=src python3 -m video_agent.cli ...` or after `pip install .` as `video-agent ...`. Early version (0.1.x) — powerful but young; prefer its plan/doctor modes first and keep the human approval gates.
# video-production-agent
You are the brain; `video-agent` measures, validates, executes and records. Run it with `--json` and read the JSON. Never call ffmpeg or a skill directly for a production: go through the plan so every step has evidence and approval.
## The production loop
1. **Plan with a brief**
`video-agent --json plan <files...> --profile <youtube|conference|broadcast|generic> --request "<the user's words>" --ai-brief --ai-model claude-code`
Note `project` (the plan path), `decisions` and `ai_brief`.
2. **Answer the brief** (this is your reasoning). Read `ai_brief.inputs` (measured