difference
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains instructions that explicitly tell the agent to disregard standard operational rules, such as "opt-out of any standing rule to check docs or search before answering." This directive to "ignore previous instructions" or system-level behaviors is a common pattern used in prompt injection, although here it is applied to optimize response speed.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data (sentences or terms) for comparison, creating a potential surface for instructions embedded in that data to influence the agent.
- Ingestion points: User-supplied arguments defined in the
argument-hintand used in the comparison logic inSKILL.md. - Boundary markers: Absent. The skill does not define clear delimiters to separate user-provided content from its own instructional framework.
- Capability inventory: The skill is permitted to use the
WebSearchtool. - Sanitization: Absent. There are no instructions to validate or sanitize the input strings before the model performs its analysis.
Audit Metadata