astropy

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs fetching and ingesting public third-party data (e.g., "Access remote FITS files (S3, HTTP)" in SKILL.md and the remote-file examples in references/fits.md, and online name queries like SkyCoord.from_name in references/coordinates.md), so the agent will read untrusted web-hosted/catalog content (FITS headers, WCS, catalog entries) that can materially influence processing and subsequent actions.