review-execution
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists only of markdown instructions for an agent to follow during a manual review process. It does not ship with any scripts, binaries, or configuration files that execute system commands or initiate network requests.
- [PROMPT_INJECTION]: The skill instructs the agent to read and process external data including project briefs, blueprints, and code diffs. This creates a surface for indirect prompt injection where malicious content in those files could attempt to influence the agent's behavior. However, the impact is minimized as the skill is focused on documentation review and lacks access to sensitive credentials or high-privilege system tools.
- Ingestion points: Processes content from <slice_path>/brief.md and <slice_path>/blueprint.md.
- Boundary markers: No specific delimiters or safety instructions are defined to isolate untrusted data.
- Capability inventory: The skill involves reading local files and updating markdown artifacts.
- Sanitization: No explicit sanitization of the ingested file content is performed.
Audit Metadata