learning-about-you
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill manages personal identification information (PII) such as names and roles to personalize interactions. It includes explicit privacy guidelines stating that sensitive data like credentials should not be stored. No evidence of unauthorized exfiltration or hardcoded secrets was found.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external codebase files to infer technical preferences, creating a surface for indirect prompt injection.
- Ingestion points: Codebase files processed via
analyze_codebase_for_profileand theReadtool. - Boundary markers: No explicit markers are used to delimit external content from instructions.
- Capability inventory: Includes file reading, shell access (Bash), and profile update tools.
- Sanitization: No evidence of sanitization or validation of data extracted from codebases.
Audit Metadata