setup-claude

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs using web searches/Stack Overflow and the Agent Browser for scraping (components/documentation-setup.md and tools_installation.md), and it supports installing and loading community skills from external sources/skills.sh or GitHub via npx (components/skills-discovery.md), which means the agent will fetch and read arbitrary third‑party user-generated content.