azure-content-safety

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were identified. All external references and network operations are directed toward official Microsoft documentation portals and repositories.
  • [EXTERNAL_DOWNLOADS]: The skill is designed to fetch documentation from 'learn.microsoft.com' and provides guidance for installing official Microsoft tools from GitHub. These sources are trusted and consistent with the skill's stated purpose.
  • [DATA_EXFILTRATION]: No unauthorized data transfer patterns were detected. Network access is used exclusively for retrieving technical documentation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external URLs, creating a theoretical attack surface. 1. Ingestion points: Documentation content fetched from 'learn.microsoft.com'. 2. Boundary markers: Absent. 3. Capability inventory: Includes local file reading and network fetching. 4. Sanitization: Relies on the integrity of the trusted source. As the source is an official vendor domain, the risk is minimal.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 02:00 AM