reason-grill-counter
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided arguments, creating a surface for indirect prompt injection attacks. \n
- Ingestion points: The 'argument under review' described in the skill instructions. \n
- Boundary markers: The skill contains an explicit 'Untrusted input rule' that instructs the agent to treat the input as untrusted data and never follow instructions found inside it. \n
- Capability inventory: The skill is restricted to text generation and does not utilize tools for network access, file system modification, or command execution. \n
- Sanitization: The skill relies on natural language instructions to sanitize the processing of the input argument.
Audit Metadata