typo3-wcag22-aa-agentic
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill follows security best practices by instructing users to manage credentials via environment variables and including manual review gates for sensitive outputs like legal accessibility statements.
- [COMMAND_EXECUTION]: The skill utilizes local shell scripts (
Skills.sh,run-a11y.sh) and standard Node.js task runners to automate project scaffolding and accessibility scanning. These operations are local to the project environment. - [EXTERNAL_DOWNLOADS]: The skill relies on standard, well-known development dependencies including Playwright, axe-core, and various utility packages (e.g.,
fast-xml-parser,handlebars). It also triggers the download of Playwright browser binaries during initial setup. - [DATA_EXFILTRATION]: The skill performs network operations to fetch sitemaps and crawl target TYPO3 website URLs for auditing purposes. These requests are restricted to the base URL configured by the user.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests and processes untrusted content from the target website's HTML, sitemaps, and accessibility audit results. This is inherent to the skill's primary function as an accessibility auditor.
Audit Metadata