wcag-audit-patterns
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains pedagogical content and standard code snippets for accessibility compliance. No evidence of obfuscation, hardcoded credentials, or unauthorized network operations was found.- [PROMPT_INJECTION]: The skill describes processes for auditing external websites, which creates a surface for indirect prompt injection. 1. Ingestion points: Auditing external URLs using browser automation (Playwright) and CLI tools. 2. Boundary markers: Absent; the provided snippets do not include instructions to the agent to ignore content from target pages. 3. Capability inventory: Browser navigation, script execution via
page.evaluate, and CLI subprocess execution vianpx. 4. Sanitization: Absent; the guide focuses on identifying accessibility violations rather than sanitizing input. This surface is typical for the intended use case of an auditor tool.
Audit Metadata