bro

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to re-process and simplify its own previous messages from the conversation history. This creates a surface for indirect prompt injection if the previous message contained untrusted data from a third party that the agent previously output.
  • Ingestion points: Processes the most recent assistant message in the conversation history (SKILL.md).
  • Boundary markers: The skill lacks explicit sanitization or delimiters for the historical content it re-processes.
  • Capability inventory: The skill explicitly forbids the use of any tools during the re-explanation process, which significantly mitigates the impact of potential injection.
  • Sanitization: No specific sanitization logic is provided for the text being simplified.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 05:12 PM