azure-active-directory

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill integrates with Microsoft Entra to process employee, department, and payroll data. This creates a standard surface for indirect prompt injection if external data fields contain malicious instructions, though the skill itself provides documentation and does not implement unsafe data handling.
  • Ingestion points: SKILL.md (via apideck.hris.employees.list example)
  • Boundary markers: None specified in this documentation-only skill.
  • Capability inventory: Network requests to vendor APIs via SDK/curl.
  • Sanitization: Not applicable to documentation instructions.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references the @apideck/unify package and fetches resources from the vendor's domains (apideck.com, specs.apideck.com) and Microsoft's official documentation. These are legitimate resources provided by the service author and platform vendor.
  • [COMMAND_EXECUTION]: The documentation includes curl command examples intended for developers to verify API coverage and interact with the Apideck Proxy API. These commands are informational and target the vendor's own infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 06:32 PM