claude-automation-recommender

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to use web search and live documentation lookups (e.g., "use web search to find recommendations" in SKILL.md and "context7
• Live documentation lookup" / Exa / Puppeteer descriptions in references/mcp-servers.md), which means the agent will fetch and read open/public third‑party content as part of its workflow.