brave-search
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. * Ingestion points: Fetches untrusted web titles and snippets from external websites via the Brave Search API. * Boundary markers: Lacks specific delimiters or instructions in the code examples to isolate untrusted search results from system prompts. * Capability inventory: Uses network requests (requests, curl) to retrieve external data for agent processing. * Sanitization: No sanitization or filtering of the fetched web content is performed before recommending its use in AI reasoning.
Audit Metadata