astro-ops
Warn
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill contains deceptive metadata, including claims of being verified in '2026-07' and references to non-existent versions such as 'Astro 7' and 'Vite 8' (SKILL.md, references/deployment.md).
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted external data through custom loaders (references/content-collections.md).
- Ingestion points: Data fetched from external API endpoints in the apiLoader and incrementalLoader functions in references/content-collections.md.
- Boundary markers: The documented code for external data ingestion does not use explicit boundary markers to isolate the fetched content.
- Capability inventory: The skill utilizes Write and Bash tools for building and deploying the site, as indicated in the allowed-tools section of SKILL.md.
- Sanitization: The author includes a security checklist in SKILL.md that recommends sanitizing user input and configuring CSP headers.
Audit Metadata