structured-data
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a Node.js script (inspect-metadata.mjs) which can launch a headless Chromium browser instance via Playwright to render and analyze web pages.\n- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes untrusted data from external websites, creating an attack surface for indirect prompt injection.\n
- Ingestion points: scripts/inspect-metadata.mjs fetches HTML, robots.txt, and sitemap content from remote URLs provided by the user.\n
- Boundary markers: The findings are structured into a report, and instructions in SKILL.md require the agent to label claims (e.g., OBSERVED, SOURCE-INDICATED), which helps contextualize the data.\n
- Capability inventory: The agent can run the inspection script and suggest modifications to the project's metadata or structured data files based on the results.\n
- Sanitization: The script parses the fetched content but does not include specific filters or sanitization logic to detect or neutralize malicious instructions embedded in the target website's content.\n- [DYNAMIC_EXECUTION]: The browser.mjs script uses dynamic import() to load the Playwright library after resolving its path at runtime based on the local environment.
Audit Metadata