exa-search

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill functions as a documentation guide for the Exa MCP server and does not contain any malicious code or instructions.
  • [EXTERNAL_DOWNLOADS]: The skill mentions downloading the exa-mcp-server package via npx. This is the official distribution method for a well-known service (Exa) and is considered safe under standard usage.
  • [CREDENTIALS_UNSAFE]: The skill correctly instructs users to handle sensitive API keys using environment variables (EXA_API_KEY) within the MCP configuration, avoiding hardcoded secrets.
  • [PROMPT_INJECTION]: The skill facilitates the retrieval of external web content, which is an inherent feature of search tools. While this introduces a surface for indirect prompt injection from external search results, there are no instructions within the skill that attempt to exploit this or bypass agent safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 01:52 PM