documentation

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

• Third-party content exposure detected (high risk: 0.80). The skill's autonomous workflow explicitly runs link-checkers and validates external URLs (see rules/auto-update-loop.md step 8 "lychee link check across all six targets" and references/research-sources.md instruction to "Replace any URL that returns a 404 before relying on a quote"), which causes the agent/tooling to fetch and interpret public third‑party web content that can influence whether updates are applied — exposing it to untrusted, user‑generated sources that could carry indirect prompt injection.