All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Initial release of n8n-nodes-openguardrails
- Check Content operation for general content safety checks
- Input Moderation operation for user input protection
- Output Moderation operation for AI response safety
- Conversation Check operation for multi-turn conversation analysis
- Configurable detection options:
- Security check (prompt attacks)
- Compliance check (content safety)
- Data security check (data leaks)
- Three action strategies for risky content:
- Continue with warning
- Stop workflow
- Use safe response
- User ID tracking for ban policy enforcement
- Comprehensive error handling and retry logic
- Custom API URL support for self-hosted OpenGuardrails
- Node icon and visual design
- Complete documentation and examples
- API key stored as password type (not visible in plain text)
- Secure credential testing with OpenGuardrails API
- Safe error messages (no API key leakage)
- Multimodal support (image + text detection)
- Webhook integration for async detection
- Batch detection API support
- Custom threshold configuration per workflow
- Statistics and reporting features
- Response template customization
- Knowledge base query operation
- Whitelist/blacklist management integration
- 1.0.0 - Initial public release
For detailed changes, see GitHub Releases.