algolia-search
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and display of external data from Algolia search indices, which is a potential surface for indirect prompt injection if the indexed content is malicious.
- Ingestion points: Search results (hits) containing user-controlled attributes like
nameanddescriptionare rendered in UI components (e.g.,ProductSearchandAutocompleteinSKILL.md). - Boundary markers: The skill does not explicitly define prompt boundaries, as it focuses on UI implementation patterns rather than direct agent instructions.
- Capability inventory: The skill enables network operations to Algolia's official domains and allows rendering search data within the application context.
- Sanitization: The implementation patterns use standard React JSX interpolation (e.g.,
{hit.name}), which provides default protection against XSS and basic content escaping by treating interpolated values as text.
Audit Metadata