brave-search-api
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes
curlto interact with the Brave Search API andjqto parse the resulting JSON data. - [EXTERNAL_DOWNLOADS]: Fetches search result data from the official Brave Search API (api.search.brave.com).
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. Ingestion points: External search results are retrieved via
curland integrated into the agent's context. Boundary markers: The skill lacks specific instructions or delimiters to isolate search results from the agent's core instructions. Capability inventory: The skill employscurlfor network operations andjqfor data processing. Sanitization: No sanitization or content filtering is performed on search result data before it is presented to the agent.
Audit Metadata