Grain MCP Server
Enables browsing, retrieving, and downloading meeting recordings and transcripts from Grain via browser automation, bypassing the need for enterprise API access.
README
Grain MCP Server
A Model Context Protocol server for Grain, a service that records and transcribes meetings.
This server provides integration with Grain through MCP, allowing users to access Grain functionality without needing the enterprise API (which is only available on enterprise subscriptions) or paid integrations like Zapier. The service is based on Playwright for browser automation.
Installation
Manual Installation
Add the Grain MCP server configuration to your MCP client:
{
"mcpServers": {
"grain_uvx": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/eadm/grain-mcp-server",
"grain-mcp-server",
"--user-data-dir",
"<absolute-path-to-browser-session-data>"
]
}
}
}
Replace <absolute-path-to-browser-session-data> with the absolute path where you want to store the browser session data. On first MCP usage, you will need to login to Grain via the browser.
Components
Tools
-
get_all_meetings: Retrieve all meetings from Grain- Returns a list of dictionaries containing meeting information, including:
id(string): Meeting IDtitle(string): Meeting titleurl(string): URL to access the meetingdate(string): Meeting date in ISO format
- Returns a list of dictionaries containing meeting information, including:
-
download_meeting_transcript: Download a meeting transcript- Required inputs:
absolute_save_path(string): The file path where the transcript will be savedmeeting_id(string): The unique identifier of the meetingtranscription_type(string): The format of the transcript file ("vtt" or "srt")
- Returns:
bool: True if the download was successful, False otherwise
- Required inputs:
Usage Examples
Some example prompts you can use with your MCP client to interact with Grain:
-
"Show me all my recent meetings" → execute the
get_all_meetingstool to retrieve a list of all meetings stored in Grain -
"Download the transcript for my last team meeting" → use
get_all_meetingsto find the meeting, then usedownload_meeting_transcriptto download its transcript
Development
- Install dependencies:
uv sync
- Run the application:
uv run grain-mcp-server
- Run with debug mode:
uv run grain-mcp-server --debug
- Run tests:
uv run pytest
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。
