mcp-skill-hardened

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly wraps the public Exa MCP (https://mcp.exa.ai/mcp) and exposes web-search and crawling tools (web_search_exa, web_search_advanced_exa, crawling_exa, company_research_exa, linkedin_search_exa) in SKILL.md and SAFETY.md, which clearly ingest and have the agent read/interact with untrusted public web/social content that could carry indirect prompt-injection.