exa-search

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The configuration instructions in SKILL.md describe using npx to run exa-mcp-server, which fetches the package from the npm registry. This is the official and standard deployment method for the Exa search MCP server.
  • [PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from the internet via search and crawling tools, creating a surface for indirect prompt injection.
  • Ingestion points: Data retrieved via web_search_exa, get_code_context_exa, and crawling_exa in SKILL.md.
  • Boundary markers: None present in the skill instructions to separate retrieved content from instructions.
  • Capability inventory: The skill utilizes search and extraction tools. No dangerous local capabilities like file writes or arbitrary shell execution were found in the skill files.
  • Sanitization: No sanitization logic for external content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 12:58 PM
Security Audit — agent-trust-hub — exa-search