astro

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to download and install the Astro framework and various integrations using package managers like npm and bun. It also contains a maintenance script, scripts/check-updates.py, which fetches version information from registry.npmjs.org and documentation manifests from docs.astro.build. These operations target official, well-known service domains for the framework and are documented for project setup and maintenance purposes.
  • [COMMAND_EXECUTION]: The documentation provides standard command-line instructions for the Astro CLI, such as astro dev, astro build, and astro add. These are expected functionalities for a developer-oriented skill supporting a web framework.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an attack surface for indirect prompt injection, as it documents features for ingesting external data via Content Collections, API endpoints, and server actions. However, it mitigates this risk by providing extensive guidance on using Zod for strict schema validation and warning users to only use raw HTML injection (set:html) for trusted content.
  • [DYNAMIC_EXECUTION]: The provided Python script scripts/check-updates.py performs dynamic network requests to verify the integrity and currency of the skill's reference content against the latest upstream versions of Astro. The script uses standard libraries and performs specific, documented tasks related to skill maintenance.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 10:30 AM
Security Audit — agent-trust-hub — astro