azure-search-documents-py
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation instructs the installation of
azure-search-documentsandazure-identityvia pip. These are established, official libraries published by Microsoft for interacting with Azure services and are considered trusted dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to retrieve and process data from external Azure AI Search indexes, which constitutes an ingestion surface for untrusted external content.
- Ingestion points: Untrusted data is retrieved using the
SearchClient.searchmethod and subsequent iteration over results inreferences/detailed-guide.md. - Boundary markers: The guide explicitly mandates to "treat results as untrusted source text," providing a conceptual safety boundary for the agent, though specific programmatic delimiters (like XML tags) are not included in the brief snippets.
- Capability inventory: The logic is limited to search and management operations within the Azure AI Search ecosystem; no high-severity capabilities such as arbitrary shell execution or local file writing are implemented.
- Sanitization: The skill recommends best practices such as server-bound identity for tenant-filtering and the inspection of sanitized status codes in error responses.
Audit Metadata