Slack MCP
A simple Model Context Protocol server for Slack integration that enables messaging, channel management, and workspace search capabilities.
README
Slack MCP Server
A Model Context Protocol (MCP) server that provides seamless Slack integration for AI assistants. This server enables AI models to interact with Slack workspaces through a comprehensive set of tools for messaging, channel management, and user interactions.
🚀 Features
Core Slack Tools
- Message Management: Send messages to channels and users
- Channel Operations: List channels and retrieve message history
- User Management: Get workspace user information
- Message Search: Search across workspace messages
- Custom Tools: Specialized tools for specific workflows
Special Features
- Smart Workday Calculation: Automatically reads messages from the previous workday
- User-Specific Filtering: Dedicated tool for reading specific user messages
- Rich Message Information: Includes user details, timestamps, and thread information
- Flexible Time Ranges: Support for custom date ranges and all-time searches
📋 Available Tools
| Tool Name | Description |
|---|---|
slack_send_message |
Send messages to channels or users |
slack_get_channels |
Get list of workspace channels |
slack_get_users |
Get list of workspace users |
slack_get_channel_history |
Retrieve channel message history |
slack_search_messages |
Search messages across workspace |
read_frontend_ladisai_messages |
Special: Read Neil Kuo's messages from frontend-ladisai channel |
🛠️ Installation
Prerequisites
- Node.js 18 or higher
- npm or yarn
- Slack Bot Token with appropriate permissions
Setup
-
Clone and install dependencies:
git clone <repository-url> cd slack-mcp npm install -
Build the project:
npm run build
⚙️ Configuration
Environment Variables
Create a .env file with the following:
# Required: Slack Bot Token from https://api.slack.com/apps
Slack Bot Permissions
Your Slack Bot needs the following OAuth scopes:
Required Scopes
channels:read- Read public channelschannels:history- Read channel message historygroups:read- Read private channelsgroups:history- Read private channel historyusers:read- Read user informationchat:write- Send messagessearch:read- Search messages
Optional Scopes
channels:manage- Create channels (if using channel creation tools)groups:write- Manage private channels
MCP Configuration
Add to your MCP client configuration (e.g., Cursor's mcp.json):
{
"mcpServers": {
"slack-mcp": {
"command": "node",
"args": [
"YOUR_PATH/slack-mcp/dist/index.js",
"--token",
"YOUR TOKEN",
"--user-id",
"YOUR PERSONAL SLACK ID"
]
}
}
}
🎯 Usage Examples
Basic Message Sending
// Send a message to a channel
{
"tool": "slack_send_message",
"parameters": {
"channel": "general",
"text": "Hello from MCP!"
}
}
// Send a direct message to a user
{
"tool": "slack_send_message",
"parameters": {
"channel": "U1234567890",
"text": "Private message"
}
}
Reading Channel History
// Get recent messages from a channel
{
"tool": "slack_get_channel_history",
"parameters": {
"channel": "frontend-ladisai",
"limit": 10
}
}
Special: Reading Neil Kuo's Messages
// Get Neil Kuo's latest messages (automatic workday detection)
{
"tool": "read_frontend_ladisai_messages",
"parameters": {
"max_results": 5
}
}
// Get all-time latest messages
{
"tool": "read_frontend_ladisai_messages",
"parameters": {
"all_time": true,
"max_results": 10
}
}
// Get messages from a specific date
{
"tool": "read_frontend_ladisai_messages",
"parameters": {
"manual_date": "2025-01-27",
"max_results": 5
}
}
User and Channel Information
// List all users
{
"tool": "slack_get_users",
"parameters": {
"limit": 100
}
}
// List all channels
{
"tool": "slack_get_channels",
"parameters": {
"exclude_archived": true
}
}
🔧 Development
Scripts
npm run build- Build the TypeScript projectnpm run dev- Build and watch for changesnpm start- Start the MCP servernpm run typecheck- Run TypeScript type checking
Project Structure
slack-mcp/
├── src/
│ ├── index.ts # Main MCP server
│ ├── slack-client.ts # Slack API client wrapper
│ └── tools/
│ ├── index.ts # Tool registry
│ ├── send-message.ts # Message sending tool
│ ├── get-channels.ts # Channel listing tool
│ ├── get-users.ts # User listing tool
│ ├── get-channel-history.ts # Channel history tool
│ ├── search-messages.ts # Message search tool
│ └── read-frontend-ladisai.ts # Custom Neil Kuo reader
├── dist/ # Compiled JavaScript
├── package.json
└── README.md
🔍 Smart Workday Logic
The read_frontend_ladisai_messages tool includes intelligent workday calculation:
- Monday → Reads Friday's messages (3 days back)
- Tuesday → Reads Monday's messages (1 day back)
- Wednesday → Reads Tuesday's messages (1 day back)
- Thursday → Reads Wednesday's messages (1 day back)
- Friday → Reads Thursday's messages (1 day back)
- Saturday → Reads Friday's messages (1 day back)
- Sunday → Reads Friday's messages (2 days back)
🐛 Troubleshooting
Common Issues
-
"Failed to get channel list"
- Check if your bot has
channels:readpermission - Ensure the bot is added to the workspace
- Check if your bot has
-
"Failed to get channel history"
- Verify the bot has
channels:historypermission - Check if the bot is invited to the specific channel
- Verify the bot has
-
"Message sent successfully" but no message appears
- Confirm the bot has
chat:writepermission - Verify the channel name/ID is correct
- Confirm the bot has
-
MCP server not responding
- Check if the build was successful:
npm run build - Verify environment variables are set correctly
- Restart your MCP client (e.g., Cursor)
- Check if the build was successful:
Testing the Server
# Test if the server responds to tool list requests
echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/list", "params": {}}' | node dist/index.js
📝 License
MIT License - see LICENSE file for details.
🤝 Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
📞 Support
For issues and questions:
- Check the troubleshooting section above
- Review Slack API documentation
- Check MCP protocol specifications
Note: This MCP server is optimized for the frontend-ladisai channel workflow and includes specialized tools for reading Neil Kuo's messages with smart workday detection.
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。