knowledge-review
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust, manual-style audit process for knowledge artifacts. It does not automate dangerous operations or request excessive privileges.
- [SAFE]: Instructions for 'Adversarial Review Mindset' and 'Anti-Bias Gate' are designed to improve the accuracy and objectivity of AI-generated content rather than bypass safety filters.
- [SAFE]: External references are limited to project-internal templates (e.g.,
.claude/templates/) and standardized protocol files (e.g.,SYNC:core-engineering-principles). - [SAFE]: Network operations described (e.g.,
web-research) are intended for evidence gathering and specifically mandate citing sources, which is consistent with the skill's primary purpose. - [SAFE]: Data processing is restricted to local files provided by the user or workflow, with no evidence of unauthorized exfiltration or exposure of sensitive system files.
Audit Metadata