exa-search
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [NO_CODE]: The provided file contains documentation and metadata but does not include any executable scripts or local code logic.
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to install a package from a third-party GitHub repository (BenedictKing/exa-search) which is not a verified vendor repository.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is fetching and processing external search results, which creates a surface for indirect prompt injection. 1. Ingestion points: Data from external web searches via Exa API is added to the agent context. 2. Boundary markers: No isolation delimiters or instructions to ignore embedded commands are specified in the instructions. 3. Capability inventory: The skill enables the agent to read and process arbitrary data from the web, though no specific tool scripts are provided in this file. 4. Sanitization: There is no evidence of filtering or sanitization logic for content retrieved from external sources.
Audit Metadata