researching-codebase

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Research External Context" (Phase 2) explicitly directs the agent to use WebFetch or a "web-researcher" subagent to fetch and read external web pages (e.g., library docs or GitHub issues), which are untrusted third-party sources whose content the agent will interpret and use to influence its findings and actions.