exa-search

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

• Third-party content exposure detected (high risk: 0.95). The skill uses an Exa MCP server to perform runtime web/code searches (e.g., web_search_exa / get_code_context_exa), which will fetch and ingest outsider-authored public web pages or repository/docs text into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The MCP config runs "npx -y exa-mcp-server" at runtime, which downloads and executes the exa-mcp-server package from the npm registry (e.g. https://registry.npmjs.org/exa-mcp-server) as a required dependency, so remote code is fetched and executed during skill runtime.