miniproject

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Research stage explicitly instructs the agent to "use_skill(brave_search) with its search and content extraction scripts" and fallback to using the lynx CLI to manually search and extract web content, so the agent will fetch and read public, user-generated/third‑party webpages as part of its required workflow, which can materially influence decisions and tool use.