All software
2 software items. Page 1 of 1. 12 items per page. 1 filters active.
12
1-2 of 2
little-canary
Prompt-injection risk sensor that reads attacks by their effect on a powerless sacrificial canary model, returning block, flag, or pass with explicit coverage state.
- AI reliability
- behavioral probe
- LLM security
- + 2
- HTML
- JavaScript
- Python
1
0
docker sandbox templates
Enhanced docker sandbox templates for using AI agents in a secure vm-based sandbox
- AI agent
- docker
- LLM security
- + 2
- Dockerfile
- HCL
- Makefile
- + 1
1
0