After Effects MCP
Connects your AI assistant to afterAI, a hosted After Effects pipeline. Edit raw clips into finished short videos with upscaling, color grade, subtitles, zoom, and beat-synced effects.
README
After Effects MCP
An MCP server that edits video in After Effects — from Claude, Cursor, or any MCP client.
after-effects-mcp connects your AI assistant to afterAI, a hosted After Effects pipeline. Send a raw clip and get back a finished, ready-to-post short video: AI upscaling (Topaz), a cinematic color grade, animated subtitles, smooth zoom effects, and beat-synced phonk edits — all rendered in After Effects with paid-tier plugins, on afterAI's machines. No After Effects install, no plugins, no render farm on your side.
Built for creators automating YouTube Shorts, TikTok and Reels. If you found this searching for an "After Effects MCP", this is the fastest way to get real AE edits from an AI agent.
What it does
| Tool | What it does |
|---|---|
edit_video |
Submit a raw clip for a full After Effects edit (upscale, color grade, subtitles, zoom, phonk edit). Returns a job id + tracking URL. |
check_status |
Follow a render's progress and get the download link when it's done. |
afterai_info |
What afterAI does, the edit options, and how to get a key. |
The finished video is emailed to the account that owns the API key, and is downloadable from the tracking URL.
Quick start
1. Get an API key
- Subscribe to a plan: https://getafterai.eu/#pricing
- Create a key: https://getafterai.eu/api-access
The API is plan-gated: each edit_video call spends one credit. If a render fails, the credit is auto-refunded.
2. Install the server
git clone https://github.com/borishalachev1/after-effects-mcp.git
cd after-effects-mcp
npm install
npm run build
This produces dist/index.js, which your MCP client runs.
3. Add it to your MCP client
Claude Code (use the absolute path to the built file):
claude mcp add after-effects -e AFTERAI_API_KEY=ak_your_key -- node /absolute/path/to/after-effects-mcp/dist/index.js
Claude Desktop / Cursor — add to your MCP config (claude_desktop_config.json or .cursor/mcp.json):
{
"mcpServers": {
"after-effects": {
"command": "node",
"args": ["/absolute/path/to/after-effects-mcp/dist/index.js"],
"env": {
"AFTERAI_API_KEY": "ak_your_key"
}
}
}
}
Coming soon: a published npm package so you can run it with
npx -y after-effects-mcpinstead of cloning.
4. Use it
"Edit this clip with a cinematic color grade and subtitles: https://drive.google.com/…"
Claude calls edit_video, then you can ask it to check_status until the download link appears.
Configuration
| Env var | Required | Default | Description |
|---|---|---|---|
AFTERAI_API_KEY |
yes (for edit_video) |
— | Your afterAI key from /api-access. |
AFTERAI_BASE_URL |
no | https://getafterai.eu |
Override the API base URL. |
edit_video options
| Option | Type | Notes |
|---|---|---|
videoUrl |
string (required) | Public, direct link to the raw clip. |
colorCorrection |
boolean | Cinematic color grade. |
colorCorrectionStyle |
string | Style id, e.g. "1". |
subtitles |
boolean | Animated subtitles. |
subtitleStyle |
string | Style id, e.g. "3". |
spokenLanguage |
string | e.g. "en", "bg". |
subtitleLanguage |
string | "same" or a code like "en". |
smoothZoom |
boolean | Smooth zoom intro. |
fullEdit |
boolean | Beat-synced phonk edit with a beat-drop freeze climax. |
watermark |
boolean | Burn in a watermark. |
watermarkText |
string | Watermark text. |
notes |
string | Free-text instructions. |
Run from source
git clone https://github.com/borishalachev1/after-effects-mcp.git
cd after-effects-mcp
npm install
npm run build
AFTERAI_API_KEY=ak_your_key node dist/index.js
How it works
This MCP server is a thin client over afterAI's public automation API (POST /api/v1/order). afterAI runs the actual After Effects pipeline (Topaz upscaling, color grade, Whisper subtitles, beat detection, render) and delivers the finished file.
License
MIT © afterAI
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。