reviewing-feature-spec
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructional logic for an AI agent to perform a peer-review task. It uses standard markdown and YAML frontmatter to define its workflow.
- [SAFE]: No external network operations, script executions, or sensitive file access patterns are present. The referenced files
implementability-bar.mdandsources.mdcontain documentation and research provenance rather than executable code. - [SAFE]: The skill instructions do not attempt to override system prompts or bypass safety filters. Phrases like 'HARD RULES' are scoped to the review workflow (e.g., 'Emit exactly one verdict line') and do not constitute prompt injection.
Audit Metadata