exa-search
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data retrieved via neural search results, which could contain malicious instructions meant to influence the agent.
- Ingestion points: Results from
web_search_exaandget_code_context_exatools described inSKILL.md. - Boundary markers: None specified in the instructions to separate search content from system instructions.
- Capability inventory: The skill enables the agent to fetch arbitrary web content and code context.
- Sanitization: No explicit sanitization or filtering of the retrieved content is mentioned.
- [EXTERNAL_DOWNLOADS]: The configuration instructions include the use of
npx -y exa-mcp-server, which downloads and executes the MCP server package from the npm registry at runtime. - Evidence: Found in the MCP Requirement section of
SKILL.md.
Audit Metadata