Quick Start Guide¶
Get the AI Ops App up and running in 5 minutes!
Prerequisites¶
- Nautobot 2.4.22+ installed
- Azure OpenAI subscription with a deployed model
- Redis server accessible
- Admin access to Nautobot
Step 1: Install the App (2 minutes)¶
# Install from PyPI
pip install nautobot-ai-ops
# Add to nautobot_config.py
PLUGINS = ["ai_ops"]
# Run migrations
nautobot-server post_upgrade
# Restart services
sudo systemctl restart nautobot nautobot-worker
Step 2: Create a Secret (1 minute)¶
- Navigate to Secrets > Secrets in Nautobot
- Click + Add
- Fill in:
- Name:
azure_gpt4_api_key - Provider: Choose your provider (e.g., Environment Variable)
- Parameters: Configure according to provider
- Click Create
Step 3: Configure LLM Model (1 minute)¶
- Navigate to AI Platform > Configuration > LLM Models
- Click + Add
- Fill in:
- Name:
gpt-4o(your Azure deployment name) - Description:
Production GPT-4o model - Model Secret Key:
azure_gpt4_api_key - Azure Endpoint:
https://your-resource.openai.azure.com/ - API Version:
2024-02-15-preview - Is Default: ☑️ Check this box
- Temperature:
0.3 - Click Create
Step 4: Test the Chat (1 minute)¶
- Navigate to AI Platform > Chat & Assistance > AI Chat Assistant
- Type a test message:
Hello! Can you help me? - Press Enter or click Send
- You should receive a response from the AI!
Step 5: Optional - Add MCP Server¶
If you have an MCP server to connect:
- Navigate to AI Platform > Configuration > MCP Servers
- Click + Add
- Configure your server details
- Set Status to Healthy
- Click Create
The agent will automatically use tools from this server!
That's It!¶
You're now ready to use the AI Ops App. Here's what you can do next:
Explore Features¶
- Try asking complex questions
- Test multi-turn conversations
- Explore the REST API
Learn More¶
- User Guide - Complete feature overview
- Use Cases - Real-world examples
- FAQ - Common questions answered
Configure More¶
- Add additional LLM models for different use cases
- Connect more MCP servers for extended capabilities
- Schedule the cleanup job for maintenance
Quick Troubleshooting¶
Chat not responding?¶
Check: 1. Is the LLM Model marked as default? ✓ 2. Does the Secret exist and have the correct API key? ✓ 3. Is the Azure endpoint accessible? ✓ 4. Check Nautobot logs for errors
Fix: Navigate to LLM Models, verify configuration, test Azure connectivity
"No LLMModel instances exist"¶
Fix: Create at least one LLM Model and mark it as default
MCP Server shows "Failed"¶
Fix:
1. Verify URL is accessible: curl https://your-mcp-server/health
2. Check firewall rules
3. Review MCP server logs
Environment Variables (Development Only)¶
For local development, you can skip Steps 2-3 and use environment variables:
# .env file
AZURE_OPENAI_API_KEY=your-api-key
AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
AZURE_OPENAI_DEPLOYMENT_NAME=gpt-4o
AZURE_OPENAI_API_VERSION=2024-02-15-preview
The app automatically detects LAB environment and uses these variables.
Production Checklist¶
Before going to production:
- [ ] Store API keys in Nautobot Secrets (not environment variables)
- [ ] Configure Redis properly in
nautobot_config.py - [ ] Set up multiple LLM models for redundancy
- [ ] Configure MCP servers with health checks
- [ ] Schedule the cleanup job (weekly recommended)
- [ ] Test error scenarios (invalid API key, network failures)
- [ ] Review security settings and permissions
- [ ] Set up monitoring for Redis and PostgreSQL
Getting Help¶
- Documentation: Full User Guide
- FAQ: Frequently Asked Questions
- Issues: GitHub Issues
- Contact: See Authors and Maintainers
Next Steps¶
Now that you're up and running:
- Explore Use Cases: Check Use Cases for examples
- Configure API Access: See External Interactions
- Learn the Architecture: Review Architecture Overview
- Extend the App: Read Extending Guide
Happy chatting! 🤖