bahakizil/Slack_MCP_Server

🚀 Slack MCP Server

A modern FastMCP server for Slack integration, built with FastMCP 2.8.0 and designed to work seamlessly with Claude Desktop and other MCP clients.

✨ Features

• 🔧 16 Powerful Tools: 6 Slack + 3 AI + 7 MCP Server Management tools
• 🤖 AI Integration: OpenAI GPT-4o-mini for intelligent responses
• 💬 Smart Slack Bot: AI can respond directly in Slack channels
• 📊 Conversation Analysis: AI-powered analysis of Slack conversations
• 🔗 MCP Server Integration: Connect to external MCP servers dynamically
• 🛠️ External Tool Access: Use tools from other MCP servers through our server
• 🧠 AI + External Tools: AI can suggest and use external MCP tools
• 🏠 Local Development: Easy setup with environment variables
• ☁️ AWS Ready: Optional AWS Secrets Manager integration for production
• 🤖 Claude Desktop: Pre-configured for Claude Desktop integration
• 🔄 Modern FastMCP: Built with latest FastMCP 2.8.0 framework
• 🛡️ Secure: Environment-based credential management

🛠️ Available Tools

📱 Slack Tools

1. send_slack_message - Send messages to Slack channels
2. get_slack_channels - List all available channels
3. get_slack_messages - Retrieve messages from channels
4. get_slack_user_info - Get user information
5. search_slack_messages - Search across workspace messages
6. create_slack_channel - Create new channels

🤖 AI-Powered Tools

7. ask_ai_question - Ask questions to OpenAI GPT model
8. send_ai_response_to_slack - AI responds directly in Slack channels
9. analyze_slack_conversation - AI analysis of Slack conversations (summary, sentiment, action items, key topics)

🔗 MCP Server Management Tools

10. add_mcp_server - Add external MCP server connections
11. list_mcp_servers - List all configured MCP servers and their status
12. connect_mcp_server - Connect to an external MCP server
13. disconnect_mcp_server - Disconnect from an MCP server
14. list_external_tools - List all tools available from connected MCP servers
15. call_external_tool - Execute tools from external MCP servers
16. ask_ai_with_external_tools - AI with access to external MCP tools

🚀 Quick Start (Local Development)

1. Setup Slack App

1. Go to Slack API and create a new app

2. Enable these Bot Token Scopes:

   • channels:read - View basic information about public channels
   • channels:history - View messages in public channels
   • chat:write - Send messages as the bot
   • groups:read - View basic information about private channels
   • groups:history - View messages in private channels
   • im:read - View basic information about direct messages
   • im:history - View messages in direct messages
   • users:read - View people in the workspace
   • search:read - Search messages (requires special approval)
   • channels:manage - Create and manage channels

3. Install the app to your workspace

4. Copy the Bot User OAuth Token (starts with xoxb-)

2. Setup OpenAI API

1. Go to OpenAI Platform
2. Create a new API key
3. Copy the API key (starts with sk-)

3. Install Dependencies

# Clone or download this repository
cd mcpslack

# Install Python dependencies
pip install -r requirements.txt

4. Configure Environment

# Copy the example environment file
cp .env.example .env

# Edit .env file and add your credentials:
# SLACK_BOT_TOKEN=xoxb-your-actual-bot-token-here
# OPENAI_API_KEY=sk-your-actual-openai-api-key-here

5. Test the Server

# Run the server directly
python main.py

# The server will start and show available tools

🤖 Claude Desktop Integration

1. Install for Claude Desktop

Add this configuration to your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "slack-mcp-server": {
      "command": "python",
      "args": ["/full/path/to/mcpslack/main.py"],
      "env": {
        "SLACK_BOT_TOKEN": "xoxb-your-actual-bot-token-here",
        "OPENAI_API_KEY": "sk-your-actual-openai-api-key-here"
      }
    }
  }
}

2. Update Path

Replace /full/path/to/mcpslack/main.py with the actual path to your main.py file.

3. Restart Claude Desktop

Close and reopen Claude Desktop. You should see all 9 tools available in Claude:

• 6 Slack tools for workspace management
• 3 AI tools for intelligent responses and analysis

🧠 AI-Powered Usage Examples

Direct AI Chat

Ask: "What are the key benefits of using microservices architecture?"
Tool: ask_ai_question

AI Slack Bot

Send AI response directly to #general channel:
Tool: send_ai_response_to_slack
Channel: general
Question: "Can you explain our new deployment process?"

Conversation Analysis

Analyze recent discussion in #planning:
Tool: analyze_slack_conversation  
Channel: planning
Analysis Type: action_items

Use Cases:

• 💼 Meeting Summaries: Analyze channel conversations for key decisions
• 📊 Sentiment Analysis: Check team mood and engagement
• ✅ Action Items: Extract tasks and follow-ups from discussions
• 🤖 Smart Responses: AI bot answers team questions in real-time

🔗 MCP Server Integration

Add External MCP Servers

Add a weather MCP server:
Tool: add_mcp_server
Server Name: weather-server
URL/Command: https://weather-mcp.example.com/mcp

Connect and Use External Tools

1. Connect to the server:
Tool: connect_mcp_server
Server Name: weather-server

2. List available tools:
Tool: list_external_tools

3. Use external tool:
Tool: call_external_tool
Server: weather-server
Tool: get_weather
Arguments: {"city": "Istanbul"}

AI with External Tools

Ask AI with access to all connected tools:
Tool: ask_ai_with_external_tools
Question: "What's the weather like in Istanbul and should I schedule an outdoor meeting?"

MCP Integration Benefits:

• 🔌 Modular: Add any MCP-compatible tool/service
• 🤝 Collaborative: Multiple team members can use shared MCP servers
• 🧠 AI-Enhanced: AI can suggest and use appropriate external tools
• 📈 Scalable: Connect to dozens of specialized MCP servers

☁️ AWS Production Deployment

Prerequisites

• AWS CLI configured
• ECR repository created
• ECS cluster and service configured
• AWS Secrets Manager secrets created

1. Create AWS Secrets

# Create Slack Bot Token secret
aws secretsmanager create-secret \
    --name slack-mcp/slack-bot-token \
    --description "Slack Bot Token for MCP Server" \
    --secret-string "xoxb-your-bot-token-here"

# Create OpenAI API Key secret
aws secretsmanager create-secret \
    --name slack-mcp/openai-api-key \
    --description "OpenAI API Key for AI features" \
    --secret-string "sk-your-openai-api-key-here"

# Create other secrets as needed
aws secretsmanager create-secret \
    --name slack-mcp/slack-app-token \
    --description "Slack App Token" \
    --secret-string "xapp-your-app-token-here"

2. Update Environment for AWS

# Set USE_AWS_SECRETS=true in your deployment environment
export USE_AWS_SECRETS=true
export AWS_REGION=us-east-1

3. Deploy with Docker

# Build and tag the image
docker build -t slack-mcp-server:latest .

# Tag for ECR (replace with your account ID)
docker tag slack-mcp-server:latest \
    123456789012.dkr.ecr.us-east-1.amazonaws.com/slack-mcp-server:latest

# Login to ECR
aws ecr get-login-password --region us-east-1 | \
    docker login --username AWS --password-stdin \
    123456789012.dkr.ecr.us-east-1.amazonaws.com

# Push to ECR
docker push 123456789012.dkr.ecr.us-east-1.amazonaws.com/slack-mcp-server:latest

# Deploy to ECS (update service)
aws ecs update-service \
    --cluster slack-mcp-cluster \
    --service slack-mcp-service \
    --force-new-deployment

4. ECS Task Definition

Use the provided task-definition.json file with your AWS account details:

# Register the task definition
aws ecs register-task-definition \
    --cli-input-json file://task-definition.json

5. Load Balancer Setup

The server will be available at your ALB endpoint:

• Health check: GET /health
• MCP endpoint: POST /mcp/

🔧 Development

Running Tests

# Install development dependencies
pip install pytest pytest-asyncio

# Run tests (when test files are created)
pytest

Code Structure

mcpslack/
├── main.py                 # FastMCP server implementation
├── requirements.txt        # Python dependencies
├── .env.example           # Environment template
├── .env                   # Your environment variables
├── Dockerfile             # Container configuration
├── task-definition.json   # ECS task definition
├── deploy.sh              # Deployment script
├── claude_desktop_config.json  # Claude Desktop config
└── README.md              # This file

🔍 Troubleshooting

Common Issues

1. "Slack client not initialized"

   • Check your SLACK_BOT_TOKEN in .env file
   • Ensure the token starts with xoxb-

2. "OpenAI client not initialized"

   • Check your OPENAI_API_KEY in .env file
   • Ensure the token starts with sk-
   • Verify you have sufficient OpenAI credits

3. "Missing scope" errors

   • Add required scopes in your Slack app settings
   • Reinstall the app to your workspace

4. Claude Desktop not loading server

   • Check the path in claude_desktop_config.json
   • Ensure Python is in your PATH
   • Restart Claude Desktop

5. AWS deployment issues

   • Verify IAM roles have correct permissions
   • Check CloudWatch logs for error details
   • Ensure secrets exist in Secrets Manager

Debugging

Enable debug mode by setting DEBUG=true in your .env file:

DEBUG=true
LOG_LEVEL=DEBUG

📋 Required Slack Permissions

Make sure your Slack app has these OAuth scopes:

• channels:read - List public channels
• channels:history - Read public channel messages
• chat:write - Send messages
• groups:read - List private channels
• groups:history - Read private channel messages
• users:read - Get user information
• search:read - Search messages (requires approval)
• channels:manage - Create channels

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test locally with Claude Desktop
5. Submit a pull request

📄 License

MIT License - see LICENSE file for details.

🙋 Support

• Check the troubleshooting section above
• Review Claude Desktop MCP documentation
• Open an issue for bugs or feature requests

Built with ❤️ using FastMCP 2.8.0