azure-diagnostics

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Command Execution for Diagnostics: The skill utilizes standard Azure CLI (az) and Kubernetes (kubectl) commands to retrieve status, health, and logs. This is aligned with the primary purpose of troubleshooting and follows established cloud administration patterns.
  • High-Privilege Debugging Capability: For advanced AKS troubleshooting, the skill includes instructions for node-level debugging using kubectl debug. This provides host-level access to cluster nodes to inspect service health and logs, which is a standard procedure for infrastructure maintenance.
  • Sensitive Data Retrieval: The workflows include commands to access application settings and connection strings (e.g., APPLICATIONINSIGHTS_CONNECTION_STRING) to verify monitoring configurations. This access is limited to the user's authenticated Azure session.
  • Indirect Prompt Injection Surface: The skill processes log data from sources like Azure Monitor and Kubernetes pods. While logs are untrusted data, the skill uses them for read-only diagnostic purposes, which is a common pattern for monitoring agents.
  • Integration with Managed Services: It leverages official Microsoft MCP tools such as AppLens and Azure Monitor for AI-powered diagnostics and telemetry analysis, ensuring that operations are performed within the vendor's supported ecosystem.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 04:31 PM
Security Audit — agent-trust-hub — azure-diagnostics