design-interface
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill interpolates user-supplied data ($ARGUMENTS) directly into the prompt without using boundary markers (such as XML tags or delimiters) or explicit instructions to treat the input strictly as data. This creates a surface for indirect prompt injection where malicious input could attempt to steer the agent's behavior.
- Ingestion points: User input enters the skill context through the $ARGUMENTS variable in SKILL.md.
- Boundary markers: The skill does not use delimiters to isolate user-provided content from the core instructions.
- Capability inventory: The instructions direct the agent to 'explore the codebase', which requires file system read and search capabilities.
- Sanitization: No validation, escaping, or filtering is applied to the user arguments before they are included in the prompt.
Audit Metadata