astro

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill contains standard educational content for the Astro framework, demonstrating legitimate project initialization and routing patterns.
  • [EXTERNAL_DOWNLOADS]: Uses standard tooling such as npm create astro@latest to initialize projects, which fetches resources from official registries.
  • [PROMPT_INJECTION]: Identifies attack surfaces for indirect prompt injection when processing untrusted data through content collections or SSR requests. The skill proactively recommends sanitizing user-supplied content before rendering with set:html. Ingestion points include src/content/ and Astro.request. Boundary markers include YAML frontmatter delimiters. Capability inventory includes file system access and network operations. Sanitization is explicitly addressed in the security notes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 04:46 AM