code-style
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by reading and reviewing project source code (e.g.,
.ts,.php,.rs,.vuefiles), which constitutes a surface for indirect prompt injection where malicious instructions could be embedded in data processed by the agent. - Ingestion points: Project source code files identified by extensions listed in
SKILL.md. - Boundary markers: None explicitly defined in the skill instructions to separate code data from agent instructions.
- Capability inventory: The skill provides instructions for coding tasks which typically involve file-system writes, shell execution, and reading existing files.
- Sanitization: No specific sanitization or filtering logic is provided within the style guidelines to handle adversarial content in reviewed code.
Audit Metadata