This project is a proxy-based monitoring system for inspecting LLM context during conversations. It provides visibility into what’s being sent to and from AI models.
Features
- Context Inspection: Monitor all prompts and responses going to LLM APIs
- Security Logging: Capture and store interaction data for security analysis
- Real-time Monitoring: Live view of active conversations
- Data Export: Export conversation logs for analysis
Architecture
The tool sits as a proxy between your application and the LLM API, intercepting all traffic for inspection.
# Example: Run the monitor
go run cmd/monitor/main.go --port 8080 --output logs/
Use Cases
- Security teams auditing AI usage
- Developers debugging LLM integrations
- Organizations needing compliance logging