Confluence Knowledge Base MCP Server

# Quick Start Guide Get your Confluence Knowledge Base running in 5 minutes! ## Prerequisites Checklist - [ ] Python 3.8+ installed (`python3 --version`) - [ ] Confluence account with API access - [ ] Gemini CLI installed (https://github.com/google-gemini/gemini-cli) ## Step 1: Get Your Confluence API Token (2 minutes) 1. Go to: https://id.atlassian.com/manage-profile/security/api-tokens 2. Click **"Create API token"** 3. Name it: "Gemini Knowledge Base" 4. **Copy the token** (you won't see it again!) 5. Keep it handy for the next step ## Step 2: Clone and Install (3 minutes) ```bash # Clone this repository git clone <your-repo-url> cd confluence-knowledge-base # Run the interactive installer ./install.sh ``` ## Step 3: Follow the Wizard The wizard will ask you: ### 1. Confluence URL ``` Enter your Confluence URL (e.g., https://yourcompany.atlassian.net): URL: https://yourcompany.atlassian.net ``` ### 2. Your Email ``` Enter your Confluence email address: Email: you@company.com ``` ### 3. API Token ``` Enter your Confluence API token: API Token: [paste the token from Step 1] ``` ### 4. Which Spaces to Index The wizard will show all your spaces: ``` Found 8 space(s): Space Key: ENG Name: Engineering Type: global Space Key: DEVOPS Name: DevOps & Infrastructure Type: global Space Key: PRODUCT Name: Product Documentation Type: global ... Which spaces do you want to include in your knowledge base? Spaces: ENG,DEVOPS ``` **Tips:** - Include: Technical, Engineering, DevOps, API docs - Skip: Personal spaces (~username), Meeting notes, HR ### 5. Build Index ``` Do you want to build the index now? (recommended) (y/n): y ``` Wait 1-2 minutes while it indexes your docs. ## Step 4: Start Using It! ```bash # Start Gemini CLI gemini ``` ``` > How does our authentication system work? > What's the deployment process for production? > Explain our API rate limits ``` **That's it!** Gemini will automatically use your documentation to answer questions. ## Verify It's Working Check that the MCP server is loaded: ``` > /mcp ``` You should see: ``` confluence-kb: Connected ✓ ``` ## What Happens Behind the Scenes 1. **Install script**: - Installs Python packages (ChromaDB, sentence-transformers, etc.) - Saves your credentials to `~/.confluence_mcp.env` - Tests connection to Confluence - Discovers available spaces 2. **Space selection**: - You choose which spaces to index - Only those spaces are downloaded 3. **Index building**: - Downloads all pages from selected spaces - Converts HTML to clean text - Splits into chunks (~1000 characters each) - Creates vector embeddings for semantic search - Stores in local ChromaDB at `~/.confluence_mcp/index/` 4. **Gemini integration**: - Adds MCP server config to `~/.gemini/settings.json` - Gemini can now call the knowledge base tools ## Common Issues ### "Python 3 not found" Install Python 3: - **macOS**: `brew install python3` - **Linux**: `sudo apt install python3 python3-pip` ### "Gemini CLI not found" Install from: https://github.com/google-gemini/gemini-cli The installer will still work - you'll just need to manually add the config to `~/.gemini/settings.json` ### "Connection failed: 401 Unauthorized" Your API token is invalid or expired. Generate a new one and run `./install.sh` again. ### "No spaces found" You might not have access to any Confluence spaces. Contact your admin. ## Next Steps ### When Documentation Changes Reindex manually: ```bash source ~/.confluence_mcp.env python3 confluence_knowledge_base.py ``` Or ask Gemini: ``` > Reindex the Confluence documentation ``` ### Set Up Automatic Weekly Updates See `REINDEXING_GUIDE.md` for cron job setup. ### Customize Configuration Edit `~/.confluence_mcp.env` to: - Add/remove spaces - Change credentials - Add custom settings Then rebuild the index. ## Example Conversations ### Architecture Questions ``` > How is our microservices architecture structured? > What databases do we use and for what purposes? > Explain our event-driven architecture ``` ### Operations Questions ``` > How do I deploy to production? > What's the rollback procedure? > Where are the deployment logs? ``` ### API Questions ``` > What are the authentication endpoints? > Show me the rate limiting rules > How do I paginate API results? ``` ## Getting Help - **Detailed setup**: See `KNOWLEDGE_BASE_SETUP.md` - **Reindexing strategies**: See `REINDEXING_GUIDE.md` - **Main README**: See `README.md` --- **You're all set!** Start asking questions about your systems and let your Confluence docs do the talking.