issue-review
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub issues (titles, bodies, and comments), creating an indirect prompt injection surface. This is effectively mitigated by a comprehensive 'Untrusted Content Rule' section.\n
- Ingestion points: Untrusted content enters the agent context via
gh issue view(Phase 1) andgh issue list(Phase 5) commands.\n - Boundary markers: The skill contains explicit instructions to ignore embedded instructions (e.g., 'ignore your rules') and provides a structured report template to maintain context boundaries.\n
- Capability inventory: The skill utilizes subprocess calls (
gh,git,bash) and writes review artifacts to the file system beneath a specific directory.\n - Sanitization: The skill requires single-quoting and escaping of all user-derived values used in shell arguments and mandates the redaction of sensitive information.\n- [DYNAMIC_EXECUTION]: The skill includes a static bash script template (Phase 4) used to sanitize the environment and verify repository integrity. This serves as a security-enhancing gate rather than a vulnerability.
Audit Metadata