Guardrails
Protect your AI applications with guardrails
Lessons (17)
1. The Guardrails Dashboard
Learn how to navigate and use the Guardrails dashboard for production configurations
2. Creating Guardrail Configurations
Step-by-step guide to creating and configuring guardrail setups
3. Injection Attack Detector
Learn how to protect your AI from malicious prompts, jailbreaks, and instruction overrides
4. PII Detector and Redaction
Detect and redact sensitive personal data including PII, secrets, and IP addresses
5. Policy Violation Detector
Enforce compliance by checking content against your uploaded policy documents
6. Anti Hallucination Detectors (adherence and relevancy)
Ensure AI responses are accurate, grounded, and relevant using Hallucination, Adherence, and Relevancy detectors
7. Viewing Detection Logs
Learn how to use logs to identify patterns, debug issues, and audit compliance
8. Configuring Deployments
Connect your guardrail configurations to AI models and create protected endpoints
9. Advanced Policy Patterns
Learn how to write effective policy rules and industry-specific examples
10. Monitoring Best Practices
Effective strategies for monitoring guardrails and maintaining secure AI systems
11. AI Proxy Integration
Integrate guardrails with minimal code changes using the AI proxy endpoint
12. Python SDK Integration
Use the Python SDK for custom workflows and direct guardrail calls
13. RAG Workflow Integration
Apply guardrails at three critical points in Retrieval-Augmented Generation pipelines
14. Quick Reference Guide
Handy reference for detector recommendations, use cases, and troubleshooting
15. Batch Processing Logs
Process multiple texts efficiently using the batch detection API with rate limit handling
16. Analyzing Guardrail Results
Parse, analyze, and generate reports from guardrail detection results
13. Understanding Guardrails Performance
Learn how to measure and optimize guardrail performance metrics