nuxt-sanity

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

• Third-party content exposure detected (high risk: 0.80). The skill’s rules repeatedly instruct the agent to fetch and render content from third-party Sanity CMS instances (e.g., useSanityQuery/useSanity in rules/core-composables.md, server fetches and validateSanityQuery in rules/core-server-routes.md, and Portable Text rendering in rules/features-sanity-content.md), which are untrusted/user-generated sources that the agent is expected to read and that can materially influence behavior (e.g., rendering, query proxying, sitemap generation, stegaClean handling).