v1.0.0
·
480 commits
to master
since this release
mcp-video v1.0.0
First stable release of the video editing MCP server for AI agents.
Highlights
- 82 MCP tools for video editing, effects, transitions, audio processing, AI features, and more
- 3 interfaces: MCP server (AI agents), Python client, rich CLI
- 690+ tests passing with real media — no mocks
- Security audited: adversarial red team audit with 11 vulnerability fixes (FFmpeg injection, path traversal, parameter bounds)
- AI-powered: Whisper transcription, Demucs stem separation, Real-ESRGAN upscaling, scene detection, silence removal
- Remotion integration: scaffold, render, and validate Remotion projects
- Audio synthesis: procedural waveform generation, presets, sequencing, effects chains
- Pro features: timeline DSL, batch processing, picture-in-picture, split-screen, chroma key, stabilization
Breaking Changes
- First stable API — no breaking changes from 0.x, but now considered stable
Dependencies
- Python >= 3.11
- FFmpeg (required)
- Optional: openai-whisper, demucs, realesrgan, opencv-python for AI features