exa-web-search
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a sophisticated orchestration pattern for search tasks. It utilizes specific tools (
web_search_exa,web_fetch_exa) to interact with the Exa MCP server athttps://mcp.exa.ai/mcp. All referenced domains (exa.ai, auth.exa.ai, dashboard.exa.ai) are official and consistent with the skill's stated purpose. - [SAFE]: The skill uses a modular architecture with reference files for extraction, filtering, and synthesis, which promotes clean context management and prevents data bloat. It explicitly instructs the agent to use subagents for bulk processing to isolate tokens and maintain clarity.
- [SAFE]: Security best practices are followed regarding authentication, recommending OAuth and guiding users to official dashboards for API keys rather than hardcoding credentials. It also includes specific guidance on date calculations to ensure accuracy for temporal queries.
Audit Metadata