MCP 服务器

Lazynext MCP Server

Enables AI agents to read, edit, and export real video projects with a full multitrack timeline, effects, captions, and export capabilities.

README

<h1 align="center">Lazynext</h1>

<p align="center"> <strong>AI video editor — local-first, agent-native, MCP-powered</strong> </p>

<p align="center"> Let AI agents read, edit, and export real video projects that remain fully editable. </p>

What is Lazynext?

Lazynext is a local-first, agent-native AI video editor. It brings conversational AI agents and professional timeline editing into the same workspace. Every edit is written to real tracks, clips, transitions, captions, effects, and media inside the project. You can continue editing manually, undo or redo changes, save versions, or hand the project to another agent.

Lazynext = local video projects + multitrack timeline + AI agents + MCP + production-ready exports.

Key features

Agent-native: the built-in agent and external MCP agents share the same editing tools.
Real timeline: multiple video and audio tracks, transitions, effects, LUTs, zooms, and keyframes.
Transcript-driven editing: word-level transcription, text-based cuts, pause handling, speakers, and linked captions.
Generation and media: images, video, speech, music, sound effects, and online media search.
Motion Graphics and WebGL: editable motion templates, custom shaders, visual effects, and transitions.
Visual geometry: in-browser person segmentation and face-safe zones — captions avoid the speaker automatically, reframe follows the subject, and overlay graphics land in empty space.
Production-ready exports: MP4, audio, captions, FCPXML, and complete project data.
Local-first: projects and media stay on your machine by default, while API keys remain server-side.

Quick start

Prerequisites

Node.js 24 (use fnm or nvm to install)
cmake (for building whisper.cpp from source on macOS)

Install and run

git clone https://github.com/Lazynext-Platform/Lazynext.git
cd Lazynext
npm ci
cp .env.example .env
npm run dev

The editor opens at http://localhost:5199.

Configuration

Edit .env to add API keys for the features you want to enable:

AI Chat: LLM_ANTHROPIC_API_KEY, LLM_OPENAI_API_KEY, or any supported provider
Generation: IMAGE_API_KEY, ELEVENLABS_API_KEY, MUREKA_API_KEY, SEEDANCE_API_KEY
Stock media: PEXELS_API_KEY, PIXABAY_API_KEY, UNSPLASH_ACCESS_KEY
Cloud transcription: ASSEMBLYAI_API_KEY

The editor runs without any API keys — you just won't have AI chat, generation, or stock media features until you add them.

Desktop app

Build native installers for macOS, Windows, and Linux:

npm run desktop:dist          # macOS arm64
npm run desktop:dist:mac-x64  # macOS x64
npm run desktop:dist:win      # Windows
npm run desktop:dist:linux    # Linux

Agent / MCP

Lazynext exposes a Streamable HTTP MCP endpoint at http://localhost:5199/api/external-mcp/mcp.

Connect external agents like Codex or Claude Code:

Start Lazynext and open Settings → MCP to copy the bearer token.
Register the MCP server in your agent client.
Use lazynext_status → list_projects → target_project → begin_edit_session → editing tools → review_edit_session.

Core capabilities

Area	Capabilities
Timeline	Multitrack editing, move, trim, split, ripple edits, snapping, keyframes, markers, undo, redo
Visuals	WebGL effects, LUTs, chroma key, zoom, transitions, custom shaders
Audio	Multiple audio tracks, sound effects, background music, voice-over, loudness, auto-ducking, vocal isolation
Transcript	Transcription jobs, word-level editing, pause compression, search, speakers, clip views
Captions	Automatic captions, named styles, translation, timeline overlays, SRT export
Motion Graphics	Built-in templates, secure sandbox, custom templates, video rendering
AI Generation	Image, video, speech, music, sound-effect jobs with progress tracking
Media	Uploads, folders, online image/video/audio search
Export	MP4, audio, captions, FCPXML, project import/export, hardware-aware H.264 acceleration
Agent	Built-in conversational agent, skills, proposal-based edits, external MCP