mcp.film / Editing & Post / VibeFrame
VibeFrame
CLI-first video workflow runtime with an MCP surface for briefs-to-renders.
What it does
VibeFrame is an agentic video workflow runtime: a coding agent turns a brief into storyboard files, generated assets, Hyperframes scenes, reviewed renders, and final MP4s. Its MCP server exposes the same engine as typed tools for project init, build, render, generation, editing, audio, detection, inspection, and timeline operations, with dry runs and cost gates for paid providers.
Connect
claude mcp add vibeframe --scope project -- npx -y @vibeframe/mcp-server
{
"mcpServers": {
"vibeframe": {
"command": "npx",
"args": [
"-y",
"@vibeframe/mcp-server"
],
"env": {
"FAL_API_KEY": "YOUR_KEY_HERE"
}
}
}
}
{
"mcpServers": {
"vibeframe": {
"command": "npx",
"args": [
"-y",
"@vibeframe/mcp-server"
],
"env": {
"FAL_API_KEY": "YOUR_KEY_HERE"
}
}
}
}
FAL_API_KEY — Provider keys are bring-your-own and only generation needs one: FAL_API_KEY (video, via Seedance), OPENAI_API_KEY (images), plus optional GOOGLE_API_KEY, RUNWAY_API_SECRET, KLING_API_KEY, XAI_API_KEY and ELEVENLABS_API_KEY per the MCP server README.
Sample tools
project_listinitbuildrendergenerate_imagegenerate_videoedit_captionaudio_duckinspect_reviewtimeline_createField notes
Correcting the last correction: the 31 Aug note said ElevenLabs wasn't involved, based on the root README alone. The MCP server's own README (packages/mcp-server) does list ELEVENLABS_API_KEY for TTS, voice-clone, dubbing and SFX, alongside OPENAI_API_KEY, ANTHROPIC_API_KEY, GOOGLE_API_KEY, XAI_API_KEY, FAL_API_KEY, RUNWAY_API_SECRET, KLING_API_KEY and an IMGBB_API_KEY or S3 upload host for image-to-video. All of them are optional. The root README's main point still holds: default narration is local Kokoro, and FFmpeg edits, scene composition, detection, timeline work and rendering all run locally for $0. Docs now link to that MCP README because it holds the full tool table. Re-checked 14 Oct 2026: the silence is over. npm @vibeframe/mcp-server is at 0.115.6 (Node >=20, bin `vibeframe-mcp`), the v0.115.6 release landed Oct 4, and a run of commits on Oct 5 moved image, TTS and transcription providers onto shared contracts and reworked dubbing. MIT, 175 stars. Pin a version, because provider wiring is visibly in motion. Requires Node 20+, FFmpeg, and Chrome/Chromium. The cost gate is real and worth using: `--max-cost` refuses to start a build whose estimate exceeds your ceiling. Claude Desktop users can skip npx entirely — the repo ships a vibeframe.mcpb extension bundle on its latest release. The project still frames the CLI as the runtime and MCP as the host surface, so shell-capable agents may prefer direct `vibe` commands for long builds.
Also in Editing & Post
The vendor's own MCP: render JSON movies, validate them, and manage the media library.
Superseded community wrapper — JSON2Video now ships its own MCP.
Local MCP and CLI toolkit for agent-driven video composition, editing, analysis, and rendering.
Hosted MCP that clips, captions, dubs, reframes, and publishes long-form footage.
AI-native Mac timeline editor with a built-in local MCP server — closed source since v0.7.6.
Cloud video editing, Studio previews, and template rendering over Shotstack's hosted MCP.
Used in
Pairs well with
Quoted dubbing, subtitles, transcription and multi-voice audiobooks from one remote MCP; a human approves each price.
Official Deepgram server: Nova transcription, diarization, and audio intelligence.
Official hosted Airtable MCP: query, create, and update production-tracking bases.