azure-ai

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (SAFE): All dependencies identified in the skill are official Microsoft Azure SDKs (e.g., azure-ai-contentsafety, @azure/search-documents). These originate from the 'microsoft' and 'azure' organizations, which are trusted per the TRUST-SCOPE-RULE.
  • [CREDENTIALS_UNSAFE] (SAFE): No hardcoded API keys or secrets were found. The SDK examples correctly use environment variables (e.g., os.environ['TRANSCRIPTION_KEY']) or placeholders like 'endpoint' and 'credential'.
  • [PROMPT_INJECTION] (SAFE): The content of SKILL.md and the SDK reference files does not contain any instructions that attempt to override agent behavior, bypass safety filters, or extract system prompts.
  • [DATA_EXFILTRATION] (SAFE): The skill does not contain any code or commands that attempt to access sensitive local file paths or send data to untrusted external endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 03:34 AM