secure-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes guidelines for reviewing untrusted code changes, which presents a surface for instructions embedded in data.
- Ingestion points: Code changes in pull requests and project source files referenced in
SKILL.md. - Boundary markers: None explicitly defined within the instruction set to delimit user code from agent instructions.
- Capability inventory: The skill body contains only text-based rules and checklists; no scripts, subprocesses, or network tools are defined or invoked by the skill itself.
- Sanitization: Not applicable as the skill provides high-level review criteria rather than automated processing code.
- [EXTERNAL_DOWNLOADS]: The skill references security policy documentation on the author's GitHub repository.
- Evidence: Links to 'github.com/Hack23/ISMS-PUBLIC' are provided for secure development, information security, and change management policy references.
Audit Metadata