KyaniteLabs/kinocut

Overview

A guardrailed video editing MCP server designed for AI agents. It uses FFmpeg and Hyperframes for processing, with a Python client and CLI. Runs locally, is fast, and is free.

Best for

Best for
Developers building AI agents that need safe, local video editing capabilities

Use cases

• Automating video edits via AI agents
• Repurposing video content with FFmpeg and Hyperframes
• Integrating video processing into agent pipelines through Python client and CLI

How to use

Install

pip install kinocut

Tools exposed

• kinocut

Tested with

Claude Desktop, Claude Code, Cursor

Example client config

{\n  "mcp_servers": [\n    {\n      "name": "io.github.KyaniteLabs/kinocut",\n      "uri": "http://localhost:8000"\n    }\n  ]\n}

Notes

A guardrailed video editing MCP server designed for AI agents. It uses FFmpeg and Hyperframes for processing, with a Python client and CLI. Runs locally, is fast, and is free.

77 stars on GitHub. Last updated 2026-07-13. Licensed Apache-2.0.

Use cases

• Automating video edits via AI agents
• Repurposing video content with FFmpeg and Hyperframes
• Integrating video processing into agent pipelines through Python client and CLI

Pros

• Guardrailed design ensures safe execution for AI agents
• Runs locally for low latency and privacy
• Free and open source with powerful FFmpeg integration

Cons

• Limited community adoption (77 stars) and documentation
• Requires familiarity with FFmpeg and Hyperframes
• Only supports Python client and CLI, no GUI

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.