enhance-prompt

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were detected. The skill operates as a text transformation tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided UI descriptions to enhance them. While this constitutes an attack surface, the risk is negligible as the skill only generates markdown text and does not execute the content.
  • Ingestion points: User input prompts processed in Step 1 and Step 3.
  • Boundary markers: The output is structured using markdown code blocks and headers to separate the enhanced prompt from other context.
  • Capability inventory: Uses Read to access DESIGN.md and Write to save next-prompt.md.
  • Sanitization: None specifically implemented for the input string, but the transformation logic acts as a structural filter.
  • [EXTERNAL_DOWNLOADS]: The skill provides a link to official documentation on stitch.withgoogle.com, which is a well-known and trusted domain.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 02:25 PM
Security Audit — agent-trust-hub — enhance-prompt