elite-audit

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to audit external web implementations, creating a surface for indirect prompt injection.\n
  • Ingestion points: The agent analyzes code or live sites ("web implementations") as part of its primary audit function in SKILL.md.\n
  • Boundary markers: No specific delimiters or safety instructions are provided to help the agent distinguish between its audit rules and potentially malicious instructions embedded in the audited content.\n
  • Capability inventory: The agent is empowered to execute CLI tools (Lighthouse, etc.) and invoke subsequent remediation skills (e.g., elite-accessibility) based on the audit results.\n
  • Sanitization: The skill does not specify procedures for sanitizing external data before processing.\n- [EXTERNAL_DOWNLOADS]: The skill recommends using well-known developer tools and performance testing services.\n
  • Evidence: References to npx lighthouse, npx unlighthouse, npx vite-bundle-visualizer, and npx source-map-explorer in SKILL.md.\n
  • Evidence: References to industry-standard services such as WebAIM, BrowserStack, and WebPageTest.\n- [NO_CODE]: This skill contains no executable scripts. It is composed entirely of Markdown documentation and reference files (e.g., critical-rules.md, pre-launch-checklist.md) used to guide the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 11:52 PM