review-spec
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a strictly defined, read-only workflow for analyzing local documents. It does not attempt to execute code or access sensitive system configuration files.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from a
spec.mdfile located atdocs/threads/*/. This constitutes an ingestion point for external content. However, the instructions explicitly define the scope as read-only and limit the agent's output to a single reporting bundle using a specific tool (/emit-pending-review), which significantly mitigates the risk of an attacker successfully overriding the agent's behavior to perform malicious actions. - [DATA_EXPOSURE_AND_EXFILTRATION]: The skill resolves paths within a specific local directory structure (
docs/threads/) and does not utilize any network-enabled commands likecurlorwget. No patterns for data exfiltration or hardcoded credentials were found.
Audit Metadata