audit-security

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is a defensive tool designed for auditing code for vulnerabilities. It correctly identifies security best practices (e.g., using parameterized queries, sanitizing HTML) and uses well-known, trusted resources like OWASP for research.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external search results and the project's source code, creating a surface for potential injection.
  • Ingestion points: The skill retrieves web search results and scrapes content via firecrawl_search and firecrawl_scrape (SKILL.md), and reads application source code for auditing.
  • Boundary markers: Absent. There are no explicit instructions to use delimiters or "ignore instructions" tags when processing external data.
  • Capability inventory: The skill is authorized to run local dependency audit tools (e.g., npm audit, pip-audit, cargo audit) and search production error logs via Sentry.
  • Sanitization: Absent. The skill does not describe any specific sanitization or validation of external content before the agent evaluates it. However, given the context of a security audit where data is analyzed rather than executed, this is considered a low-risk surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 06:52 PM
Security Audit — agent-trust-hub — audit-security