design-audit
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it systematically reads and processes data from the codebase being audited without employing safety boundaries.
- Ingestion points: The skill reads various project files, including configuration files (e.g., package.json, pubspec.yaml), UI component files (screens, widgets, pages), and the MEMORY.md file during its survey and audit phases.
- Boundary markers: The instructions lack explicit delimiters or warnings to the agent to disregard instructions that might be embedded in the files it analyzes.
- Capability inventory: The skill utilizes filesystem read access across the project and write access to MEMORY.md to store baseline scores.
- Sanitization: There is no mention of sanitizing, escaping, or filtering the content read from the files before the agent analyzes it, which could allow instructions hidden in the audited code to be executed by the agent.
Audit Metadata