Antigravity Docs MCP
Provides AI agents with up-to-date documentation for the Antigravity IDE, including a built-in scraper and tools to list and search docs.
README
Antigravity Docs MCP
A Model Context Protocol (MCP) server that provides AI agents with up-to-date documentation for the Antigravity IDE. It includes a built-in scraper to automatically fetch and update the latest docs directly into a local knowledge base.
Features
- Server: Provides
docs://indexto list pages, and toolslist_document_namesandsearch_docsto find relevant information. - Scraper & Extractor: Python scripts powered by Playwright to crawl the Antigravity Docs site and extract readable Markdown.
- Auto-Updater: A single bash script
update.shto fully automate the parsing of the site and injecting the content into theknowledge_base/folder.
Installation
1. Clone the repository
git clone https://github.com/imgt511/Antigravity_Docs_MCP.git
cd Antigravity_Docs_MCP
2. Setup the Python Environment (For the Scraper)
The scraper uses Python and Playwright. Build a local virtual environment:
python3 -m venv venv
source venv/bin/activate
pip install playwright
playwright install
3. Setup the Node Environment (For the Server)
The MCP server itself is served using the official MCP Node SDK.
npm install
Usage
Connecting the MCP Server to Antigravity IDE
Point your Antigravity IDE settings (or any other MCP-compatible client) to execute the Node.js server:
{
"mcpServers": {
"antigravity-docs": {
"command": "node",
"args": ["/absolute/path/to/Antigravity_Docs_MCP/mcp_server.js"]
}
}
}
How to Refresh the Knowledge Base
When new updates are released for the Antigravity IDE, you can easily pull the latest documentation yourself!
The Easy Way
Simply run the included update script from inside the project directory:
./update.sh
What it does:
- Runs
scraper.pyto find any new pages on the website. - Runs
extractor.pyto overwrite yourknowledge_base/folder with the newest text. - Your MCP server instantly starts using the new files. There is no need to restart anything!
Daily Catchup Log
- Update: 2026-03-13 21:03:50 - Automated Catchup push.
- Update: 2026-03-17 20:29:47 - Automated Catchup push.
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。