code-security-audit

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (user-provided code or API documentation) for security auditing. This creates a surface for indirect prompt injection where malicious instructions within the audited source could attempt to deceive the agent or influence its findings.
  • Ingestion points: User-specified files, modules, or API endpoints as described in SKILL.md and SUBAGENT.md.
  • Boundary markers: The instructions do not define specific delimiters or 'ignore instructions' warnings for the input content.
  • Capability inventory: The instructions describe synthesis and reporting tasks; no direct subprocess execution, file-system writes, or network exfiltration logic is defined in the skill markdown.
  • Sanitization: No explicit sanitization or filtering of the input content is specified before the analysis phase.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 03:06 AM