Skip to main content
Glama

Orccut

A real video editor for AI agents, served over MCP.

Orccut gives any MCP-capable agent (Claude Code, Claude Desktop, or anything else that speaks the protocol) an actual editing model — not a wrapper around one ffmpeg command. Projects are timeline snapshots with a journaled history of named operations; you and your agent work on the same projects — the agent edits over MCP, you review and tweak in the browser editor at orccut.com.

Connect your agent

  1. Sign up at orccut.com and create an API token.

  2. Point your agent at the MCP endpoint.

Claude Code:

claude mcp add --transport http orccut https://orccut.com/mcp \
  --header "Authorization: Bearer <your token>"

Any other MCP client, in its JSON config:

{
  "mcpServers": {
    "orccut": {
      "type": "http",
      "url": "https://orccut.com/mcp",
      "headers": { "Authorization": "Bearer <your token>" }
    }
  }
}

Then ask the agent for something real: “make a 30-second cut of this video with a title and captions, and export it.”

Related MCP server: ffmpeg-mcp-video-editor

What the agent gets

44 tools over one consistent model:

  • Projects & history — create/list projects, journaled operations (editor_get_history shows every edit ever made), named versions with restore, annotations.

  • Timeline — multiple video/audio tracks, clips with trim/split/move/ resize, transforms with keyframes, transitions, canvas fit, color adjustments.

  • Text & graphics — text overlays, image/video overlays, covers.

  • Audio — music beds, per-clip audio, volume envelopes.

  • Media in — media by http(s) URL, downloaded and checked server-side before it touches a timeline.

  • Understanding — media analysis (scenes, audio peaks, keyframes), auto-captions, voiceover.

  • Out — validation, preview frames, full exports with presets.

Every edit validates its arguments, returns a fresh project snapshot with a bumped version, and lands in the journal. Nothing edits in place.

Tool documentation

The skills/ directory documents the tool surface as ready-made instructions you can hand to any agent: video-editing walks through the full workflow — project → media → timeline edits → captions → validate → export — and the conventions that make agent edits reliable.

License

The documentation in this repository is released under the MIT license.

Related MCP Connectors

Related MCP Servers

  • F
    license
    Not graded
    quality
    D
    maintenance
    An MCP server that exposes FFmpeg as a structured tool set for AI agents, enabling timeline-based video editing, preview, rendering, and analysis with an optional LLM autopilot.
    -
  • A
    license
    Not graded
    quality
    C
    maintenance
    An MCP server that gives AI agents a stateful multi-track video editor powered by MLT, enabling them to assemble timelines, add media and text, apply filters and transforms, preview downscaled frames, and render final videos to disk.
    MIT