ai-sensitive-data-disclosure
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a documentation-only resource providing techniques for security professionals to test for LLM vulnerabilities. It describes manual methods for testing Cross-tenant leakage, RAG context snooping, and training data extraction. All advice is framed within ethical guidelines (e.g., 'prove impact minimally', 'use your own two tenants', 'redact records'). No executable code or automated attack patterns are present in the provided file.
Audit Metadata