runtime-verification-discipline
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected in the skill instructions. The skill focuses on improving the fidelity of software verification by encouraging AI agents to drive real or simulated environments instead of relying on human checklists.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted data source (git diff outputs and source code content via grep) to determine if UI verification logic should be triggered.
- Ingestion points: git diff output and file contents processed during the 'UI verification face precheck' (SKILL.md).
- Boundary markers: None explicitly defined for the git analysis stage.
- Capability inventory: This skill serves as a logic layer and does not perform dangerous operations itself; it delegates execution to other capability skills or project-specific scripts.
- Sanitization: None specified for the data read via git/grep. However, the logic is limited to workflow branching (determining if a screenshot is needed) and does not involve executing instructions extracted from the data.
Audit Metadata