web-content-audit
Warn
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DYNAMIC_EXECUTION]: The skill documentation describes a process of generating and operationalizing audit scripts (.mjs files). It specifically suggests using the Node.js vm module to load data from course-data.js, which involves executing code from a file at runtime. This pattern poses a risk if the content of the target file is influenced by untrusted sources.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process various untrusted data sources to generate human-readable reports.
- Ingestion points: course-data.js, markdown files in course-package/, index.html content, and dist/ebook.docx via JSZip.
- Boundary markers: No explicit boundary markers or instructions to ignore embedded content are present in the provided script logic.
- Capability inventory: The skill possesses the capability to write files to the local disk (fs.writeFile) and potentially use Playwright for visual captures.
- Sanitization: There is no evidence of sanitization or escaping of the content extracted from source files before it is included in the markdown audit reports.
- [EXTERNAL_DOWNLOADS]: The skill utilizes well-known external Node.js libraries such as jszip for inspecting document structures and playwright for capturing screenshots. These dependencies are standard for the described auditing tasks.
Audit Metadata