close-verification
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a logical workflow for financial auditing and does not contain any malicious patterns such as remote code execution, data exfiltration, or persistence mechanisms.
- [SAFE]: The skill is designed to handle sensitive financial documents and reads configuration from
../../CLAUDE.md. This access is necessary for its stated purpose and does not involve unauthorized data transmission. - [SAFE]: Although the skill ingests untrusted data from financial reports, it lacks dangerous tools or execution capabilities (capability inventory: logic only), mitigating risks associated with indirect prompt injection.
Audit Metadata