web-scanning
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for using standard command-line tools such as
npx,curl, andgrepto automate web scanning and page inventory tasks. - [EXTERNAL_DOWNLOADS]: It references well-known Node.js packages including
@axe-core/cli,capture-website-cli, andplaywright. These are industry-standard tools for accessibility testing and browser automation from reputable sources. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection via the ingestion of untrusted data from external URLs and sitemaps (Ingestion points: SKILL.md). While explicit prompt boundary markers are absent, the capabilities (npx, curl, grep) are directed toward technical auditing tasks. Sanitization is not explicitly defined in the instructions, but the specialized nature of the auditing tools (axe-core, playwright) significantly limits the risk of the agent interpreting scanned content as instructions.
Audit Metadata