loi-25-website
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill performs audits by searching through a user's codebase (scripts, forms, and HTML). This creates an attack surface where a malicious codebase could contain instructions intended to influence the agent's audit results (e.g., hidden comments instructing the agent to ignore certain compliance failures).
- Ingestion points: The skill instructs the agent to read and search through user-provided source code files.
- Boundary markers: None specified; the agent processes the files directly as text.
- Capability inventory: The agent generates a report based on its findings but does not possess high-privilege capabilities like network exfiltration or file writing in this context.
- Sanitization: No specific sanitization or escaping of the processed code is mentioned.
Audit Metadata