dramacraft

DramaCraft 是一个专业的短剧视频编辑 MCP (Model Context Protocol) 服务，集成国产中文大模型 API，实现剪映的智能自动化编辑功能。项目已完成从视频分析到草稿生成的完整解决方案

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio whatyun-dramacraft uv run dramacraft start \
  --env LLM__API_KEY="your_api_key_here" \
  --env LLM__PROVIDER="baidu" \
  --env LLM__SECRET_KEY="your_secret_key_here" \
  --env VIDEO__TEMP_DIR="./temp" \
  --env VIDEO__OUTPUT_DIR="./output" \
  --env JIANYING__INSTALLATION_PATH="/path/to/JianyingPro.app"

How to use

DramaCraft is a Python-based MCP server that orchestrates intelligent video editing workflows by integrating cutting-edge local model calls with Jianying (剪映) project handling and domestic large-language models. It exposes eight core MCP tools to automate the end-to-end editing pipeline, including generate_commentary for multi-style narration, smart_video_edit for end-to-end editing, and create_jianying_draft for draft project creation. To run the server, install the package with uv, then start the service with uv run dramacraft start. The included examples show how to configure clients (Cursor, Claude Desktop, VS Code) to connect to this MCP server, pass environment variables for LLM providers, and launch the editing tools remotely. Once running, you can trigger specific tools via the MCP interface, monitor progress in real time, and access created outputs in the configured output directories.

How to install

Prerequisites

Python 3.9+ installed on your system
uv (modern Python package manager) installed -剪映 (Jianying) installed or accessible via path for project import
Access to a Baidu LLM provider or other supported LLM integrations if you plan to use LLM features

Installation steps

Install the DramCraft package via uv or pip: uv add dramacraft

or

pip install dramacraft
Verify installation and dependencies uv run dramacraft doctor uv run dramacraft config --validate
Prepare environment variables and configuration Create a .env file or set environment variables as demonstrated in the README: LLM__PROVIDER=baidu LLM__API_KEY=your_api_key_here LLM__SECRET_KEY=your_secret_key_here JIANYING__INSTALLATION_PATH=/Applications/JianyingPro.app VIDEO__TEMP_DIR=./temp VIDEO__OUTPUT_DIR=./output
Start the MCP server uv run dramacraft start
(Optional) Debug mode or additional commands uv run dramacraft start --debug uv run dramacraft --help

Additional notes

Tips and common considerations:

Ensure Jianying Pro is installed and the installation path is correctly set in JIANYING__INSTALLATION_PATH.
The LLM provider configuration (LLM__PROVIDER, API key, secret) must be valid for generate_commentary and narrative features.
The server uses asynchronous processing; for high throughput, monitor resource usage and enable the built-in monitoring tools with uv run dramacraft monitor --realtime.
If you encounter environment variable issues in editors (Cursor, Claude Desktop, VS Code), copy the env block from the examples and ensure the hosting shell exports them before launching the MCP client.
The eight MCP tools cover end-to-end editing, from smart_video_edit to batch_process; use the monitor and config commands to validate tool availability and performance before production use.