azure-cache-redis

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill retrieves markdown documentation from learn.microsoft.com. This is an official Microsoft domain and a well-known service for technical documentation.
  • [DATA_EXFILTRATION]: No data exfiltration patterns were identified. Network access is restricted to fetching technical guidance from verified Microsoft endpoints.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external content from documentation pages. While this represents a data ingestion surface, the content is sourced from a highly trusted official domain, significantly mitigating the risk of adversarial instruction injection.
  • Ingestion points: External documentation fetched from learn.microsoft.com via mcp_microsoftdocs or fetch_webpage tools.
  • Boundary markers: None explicitly defined for external content.
  • Capability inventory: No file system writes, subprocess executions, or other dangerous capabilities found in the skill instructions.
  • Sanitization: No explicit sanitization of the fetched markdown is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 01:06 PM