create-skill

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

• Third-party content exposure detected (high risk: 0.90). The Diagnose Mode procedure explicitly instructs the agent to perform "Authoritative web search" and to "invoke WebSearch and WebFetch against authoritative sources" (see Step 6.5 / rules/diagnose-mode.md), which requires fetching and reading open/public third-party web content that can change the agent's proposed edits and actions.