# Slack Integration Setup Guide
This guide will help you connect your MCP server to Slack API to fetch and store channel messages in your RAG knowledge base.
## Prerequisites
1. **Install Slack SDK:**
```bash
cd /path-to-your-repo/kedro-mcp-rag
/opt/anaconda3/bin/pip install slack-sdk
```
## Step 1: Create a Slack App
1. **Go to Slack API:** https://api.slack.com/apps
2. **Create New App:**
- Click "Create New App" → "From scratch"
- App Name: "Kedro Knowledge Collector"
- Select your workspace
## Step 2: Configure App Permissions
1. **Go to "OAuth & Permissions"**
2. **Add Bot Token Scopes:**
```
channels:read # View basic channel info
channels:history # Read message history from public channels
groups:read # View private channel info
groups:history # Read message history from private channels
users:read # Read user information
```
3. **Install App to Workspace:**
- Click "Install to Workspace"
- Copy the **Bot User OAuth Token** (starts with `xoxb-`)
## Step 3: Set Environment Variable
Create a `.env` file or set the environment variable:
```bash
# Option 1: Create .env file
echo "SLACK_BOT_TOKEN=xoxb-your-token-here" > .env
# Option 2: Export directly
export SLACK_BOT_TOKEN="xoxb-your-token-here"
```
## Step 4: Invite Bot to Channels
1. Go to the Slack channels you want to collect data from
2. Type: `/invite @Kedro Knowledge Collector`
3. The bot needs to be in the channel to read its history
## Step 5: Use MCP Tools
Once your Claude Desktop MCP server is running, you can use these new tools:
### List Available Channels
```
Please list my Slack channels
```
### Add Channel Data to Knowledge Base
```
Please fetch the last 30 days of messages from our #kedro-support channel and add them to the knowledge base
```
### Search with Source Filtering
```
Search for "pipeline error" but only show results from Slack discussions
```
## Example Usage
1. **Discover channels:**
- The MCP server will show you all channels your bot has access to
2. **Collect support data:**
- Import messages from channels like `#kedro-help`, `#data-engineering`, etc.
- The system automatically filters for question-like messages
3. **Enhanced search:**
- Search both documentation and real user discussions
- Get context from actual user problems and solutions
## Security Notes
- **Bot Token Security:** Keep your bot token secure and don't commit it to version control
- **Channel Access:** The bot can only read channels it's invited to
- **Data Privacy:** Consider your organization's data policy before importing sensitive channels
## Troubleshooting
### Common Issues:
1. **"Slack integration not available"**
```bash
/opt/anaconda3/bin/pip install slack-sdk
```
2. **"Invalid token"**
- Check your `SLACK_BOT_TOKEN` environment variable
- Ensure the token starts with `xoxb-`
3. **"Bot not in channel"**
- Invite the bot to the channel: `/invite @your-bot-name`
4. **"No messages found"**
- Check if there are messages in the time range
- Verify bot has `channels:history` or `groups:history` permission
## Environment Variables
```bash
# Required
SLACK_BOT_TOKEN=xoxb-your-bot-token-here
# Optional - for your existing PYTHONPATH
PYTHONPATH=/Users/Huong_Nguyen/dev/kedro-mcp-rag/
```
After setup, your MCP server will be able to fetch Slack channel data and integrate it with your Kedro documentation for enhanced AI assistance!
