astro-framework

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md and references include concrete examples of fetching and ingesting external APIs and remote images (e.g., SKILL.md Quick Reference: const data = await fetch('https://api.example.com/data'); references/content-collections.md custom loader that fetches 'https://api.example.com/products'; and image remotePatterns like '**.unsplash.com'), which shows the agent is expected to read and act on untrusted third‑party content that can influence rendering and tool behavior.