mcp_slack

# Slack-MCP This project is a **Multi-Tool MCP (Model Context Protocol)** server for Slack automation using the [FastMCP](https://github.com/modelcontextprotocol/fastmcp) library. It allows you to fetch and process Slack messages using AI agents.This project is great example to learn and Build Understanding reagarding MCPs. --- ## 📂 Folder Structure ``` slack-mcp/ ├── main.py # Main MCP tool server with multiple tools ├── .env # Store your SLACK_TOKEN safely here ├── smithery.yaml # Configuration for Smithery deployment ├── README.md # This file ├── uv-lock └── pyproject.toml # Python dependencies ``` --- ## 🚀 Tools in This MCP ### 1. `get_recent_messages` Fetches the most recent messages from a given Slack channel. * **Inputs**: `channel_id`, `limit` * **Returns**: Formatted recent messages. ### 2. `get_user_messages` Fetches messages from a specific user within a channel. * **Inputs**: `channel_id`, `username`, `limit` * **Returns**: Messages sent by the user. ### 3. `summarize_recent_messages` Summarizes the last N messages from a channel using a connected LLM. * **Inputs**: `channel_id`, `limit` * **Returns**: AI-generated summary. ### 4. `group_messages_by_user` Groups recent channel messages by sender. * **Inputs**: `channel_id`, `limit` * **Returns**: A grouped format like: ``` @user1: - Message 1 - Message 2 @user2: - Message A ``` --- ## 🚪 Slack OAuth Setup & Token Generation ### 1. Create a Slack App * Go to: [https://api.slack.com/apps](https://api.slack.com/apps) * Click **Create New App** > From Scratch * Give it a name and choose your workspace ### 2. Set OAuth Scopes In **OAuth & Permissions**, under `Bot Token Scopes`, add: * `channels:read` * `channels:history` * `users:read` * `chat:write` *(optional for sending messages)* ### 3. Install to Workspace * Click **Install App to Workspace** * Authorize permissions * Copy the **Bot User OAuth Token** (starts with `xoxb-...`) ### 4. Save the Token Create a `.env` file: ```bash SLACK_TOKEN=xoxb-your-token-here ``` If you change scopes later: * Revisit **OAuth & Permissions** * Click **Reinstall App** to apply new scopes --- ## ✨ Testing the Tools ### On MCP Inspector (Local) 1. Run the server: ```bash python main.py ``` 2. Use [MCP Inspector](https://inspector.modelcontextprotocol.org/) to call tools and test output ```bash mcp dev main.py ``` ### On Smithery (Cloud) 1. Push your code to GitHub 2. Deploy your MCP to Smithery 3. Use the "Run tool" tab to test individual tools ### On Claude (Cloud Desktop) You can connect these tools via Smithery to Claude workflows for real-time Slack insights and summaries. --- ## 🚜 Requirements Install dependencies with: ```bash uv pip install -r pyproject.toml ``` Or manually: ```bash pip install fastmcp httpx python-dotenv ``` --- ## 🔧 Environment File (.env) ```bash SLACK_TOKEN=xoxb-123-your-token ``` ## 🎉 Contributing Pull requests are welcome. If you’d like to add new Slack tools (e.g., `send_message`, `track_reactions`, etc.), feel free to open an issue or PR. --- ## 📹 Video Demo On claude desktop: https://github.com/user-attachments/assets/37d905a5-6382-40ce-9500-2dd744656546 --- ## 🙏 Acknowledgments * [FastMCP](https://github.com/modelcontextprotocol/fastmcp) by Model Context Protocol * [Smithery.ai](https://smithery.ai) for deploying and testing MCPs * Slack Web API Docs: [https://api.slack.com/](https://api.slack.com/) --- ## 📊 Example Output > Grouped Slack Messages: ``` @bakar: - Just pushed latest PR - Working on the backend API @hamza: - Updated Figma designs - Fixed navbar bug ``` ## 📊 Smithery.ai (Cloud Deployment) Here is the URL go and checkout this mcp_server: [Slack_MCP_Server](https://smithery.ai/server/@Abu-BakarYasir/my_slack_mcp) --- ## 🧑‍💻 Author Built by [Abu Bakar Yasir](https://github.com/Abu-BakarYasir) — Computer Engineering @ COMSATS + AI Software Engineer | Full Stack Developer 🚀