learning

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWSAFE
Full Analysis
  • Indirect Prompt Injection (LOW): The skill possesses an attack surface for indirect prompt injection by processing and persisting untrusted user data.
  • Ingestion points: Processes user-provided themes, concepts, and study notes through memory_search and memory_save operations.
  • Boundary markers: No explicit delimiters or boundary markers are used to separate user-provided content from instructions in the memory or file storage templates.
  • Capability inventory: The skill is capable of searching and saving to memory, and writing notes to the local filesystem at data/notes/.
  • Sanitization: There is no evidence of sanitization or filtering for content stored in memory or notes.
  • Data Storage (INFO): The skill explicitly defines storage for learning records at data/notes/. This is a legitimate functional behavior for a knowledge management tool but should be monitored for sensitive data accumulation.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 08:51 AM