azure-search-documents-py

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [Secure Authentication Practices]: The skill recommends using modern identity-based authentication methods over static API keys. This approach enables secure, keyless authentication via managed identities and environment-based configuration, reducing the risk of credential exposure.
  • [Proper Resource Lifecycle Management]: Code examples utilize context managers for client instances. This ensures that network transports and authentication caches are deterministically closed, preventing resource leaks.
  • [Data Ingestion Surface]: The skill facilitates the retrieval of documents from an external search index, creating a surface where untrusted data enters the agent's context. This is the intended functionality for retrieval-augmented tasks but requires awareness of potential data poisoning.
  • Ingestion points: SearchClient.search in SKILL.md and KnowledgeBaseRetrievalClient.retrieve in references/agentic-retrieval.md.
  • Boundary markers: Code snippets focus on demonstrating library capabilities and do not explicitly include boundary delimiters or instructions to ignore embedded content.
  • Capability inventory: The skill provides tools for search operations, index management, and document updates, without performing high-risk actions like arbitrary command execution based on retrieved data.
  • Sanitization: The examples demonstrate processing and displaying search results without showing specific sanitization or validation logic for the content of the retrieved documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:18 PM
Security Audit — agent-trust-hub — azure-search-documents-py