Kinocut
🐍 🏠 🍎 🪟 🐧 - Guardrailed video editing for AI agents with 135 MCP tools for FFmpeg editing, captions, audio, effects, Hyperframes, repurposing, resumable workflows, quality gates, and provenance receipts. Includes a Python client and kino CLI; runs locally with no required API keys.
Quick Install
{
"mcpServers": {
"kyanitelabs-kinocut": {
"command": "npx",
"args": [
"-y",
"kyanitelabs-kinocut"
]
}
}
}Using an AI coding agent (Claude Code, Cursor, etc.)? Copy a ready-made prompt that tells it to fetch the setup instructions and install this server for you.
Documentation Overview
Kinocut
Local-first FFmpeg tools, Video Receipts, quality gates, Hyperframes, and Shorts/Reels repurposing — for Claude Code, Cursor, and any MCP client. Free, Apache-2.0. Formerly mcp-video.
Who it is for: people who need agent-repeatable local video edits with a receipt a human can approve — not a hosted editor.
What you get: CLI + MCP tools that turn a local interview or podcast into captioned vertical clips with a Video Receipt you can re-run.
Why it wins
- Repurposing with receipts — one recording → captioned Shorts/Reels/TikTok packages with manifests and review artifacts
- Podcast and interview cuts — strongest segment, audio normalize, chapters, export
- Agent-driven media in CI — repeatable edits from Claude Code, Cursor, Codex-style clients, or scripts
Try it
pip install -U kinocut
kinocut --help
Published package: 1.11.1 (2026-07-24). Install matrix and golden path live in docs/.
Proof
| Surface | Tip |
|---|---|
| PyPI / npm / GitHub Release | 1.11.1 — pip install -U kinocut |
This repository (master) | MCP + CLI surface beyond the last release; see release notes before treating tip as published |
Docs
License
Apache-2.0. See LICENSE.