verify-and-stop
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon "acceptance conditions" which are external inputs. If these inputs originate from untrusted sources (such as third-party code or documentation), they could contain instructions that manipulate the agent's verification process.
- Ingestion points: Acceptance conditions defined in the task context (referenced in
SKILL.md). - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the acceptance criteria.
- Capability inventory: The skill instructs the agent to "Run focused checks", which typically involves executing shell commands or scripts.
- Sanitization: No explicit sanitization or validation of the acceptance conditions is performed before they are used to generate checks.
Audit Metadata