update-skill-learnings

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing session content and user-provided text for persistent storage.
  • Ingestion points: The skill uses the $request argument and current session context as data sources.
  • Boundary markers: No explicit delimiters are used for the input data.
  • Capability inventory: The skill has file modification permissions (Edit, Write) to update repository documentation.
  • Sanitization: The risk is mitigated by a mandatory human-in-the-loop requirement (AskUserQuestion) for approval of all changes, and the disable-model-invocation: true setting which prevents unauthorized execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 12:24 PM
Security Audit — agent-trust-hub — update-skill-learnings