reflection

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted user input from conversation history to suggest behavior updates. * Ingestion points: Conversation history and tool usage logs. * Boundary markers: Absent; the instructions do not include delimiters to separate data from instructions. * Capability inventory: Proposes modifications to skill definitions and project configuration files (CLAUDE.md). * Sanitization: Mitigated by a mandatory user confirmation step before any changes are applied.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 06:25 PM