azure-search-documents-py
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to facilitate the ingestion and processing of data from external sources, specifically Azure AI Search indexes and Azure Blob Storage containers. This architectural pattern creates an attack surface where untrusted content could contain hidden instructions intended to influence the agent.
- Ingestion points: The
references/detailed-guide.mdfile defines methods for querying search indexes (client.search) and establishing connections to data sources (SearchIndexerDataSourceConnection). - Boundary markers: The provided documentation lacks explicit instructions or delimiters that would warn the agent to treat retrieved search content as untrusted data or to ignore embedded instructions within that content.
- Capability inventory: The skill utilizes the
azure-search-documentslibrary to perform network-based retrieval of documents and metadata. - Sanitization: No sanitization, validation, or filtering logic is present in the examples to mitigate the impact of malicious content retrieved from the search index.
- [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install the
azure-search-documentsandazure-identityPython packages via pip. These are well-known, official libraries maintained by Microsoft for interacting with Azure services.
Audit Metadata