code-smells-audit
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code (source files or git diffs) and includes it in its reasoning and reports, creating a surface for indirect prompt injection. \n
- Ingestion points: Code read via
git diffand file system access inSKILL.md(Step 1).\n - Boundary markers: The skill instructs the agent to "Verify: adversarially, against the card" (Step 3), reading candidate code as a "skeptic."\n
- Capability inventory: Uses
Bashfor git operations and writes audit reports to thedocs/audits/directory (Step 4).\n - Sanitization: No explicit sanitization or filtering of analyzed content is defined for the reporting phase.\n- [COMMAND_EXECUTION]: The skill executes system commands via the
Bashtool for target resolution and catalog validation. \n - Git commands (
git diff,git status,git merge-base) are used to determine the audit scope.\n - The skill includes an internal validation script
scripts/validate-cards.shthat utilizes standard Unix utilities to perform structural checks on the reference material.
Audit Metadata