hrt-security-pass
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input as part of its primary security auditing function.
- Ingestion points: The skill reads external project files including
specs,designdocuments, and code diffs to determine if trust boundaries are touched. - Boundary markers: The instructions lack explicit delimitation or warnings for the agent to ignore instructions embedded within the analyzed content, creating a risk that malicious text in a code diff could override the auditor's logic.
- Capability inventory: The skill is authorized to write security findings back to project files (
design.md,specs). An injection could lead to the agent erroneously skipping a security pass or failing to report critical vulnerabilities. - Sanitization: No sanitization or validation of the input text is performed before the agent processes it for threat modeling.
Audit Metadata