company-cfo
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs shell commands for git-based report management (git add, git commit, git push) and GitHub CLI operations (gh pr create). These commands are used legitimately to manage the lifecycle of financial reports.
- [PROMPT_INJECTION]: The skill processes data from external financial APIs and local configuration files, which constitutes a surface for indirect prompt injection. Ingestion points: raw data from bank, payroll, and payment APIs (e.g., Stripe, Gusto) and CLAUDE.md configuration. Boundary markers: none defined. Capability inventory: shell execution (git/gh) and file system writes. Sanitization: none specified. This risk is inherent to the data-processing nature of the skill and is not considered a malicious finding.
Audit Metadata