azure-compute

Fail

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: Hardcoded passwords are used in SKILL.md command examples.\n
  • Evidence: The instruction file contains literal credentials such as --admin-password "P@ssw0rd123!" and DB_PASSWORD=s3cr3t.\n- [COMMAND_EXECUTION]: The skill documents commands that facilitate arbitrary script execution on remote infrastructure.\n
  • Evidence: The az vm run-command invoke command is included, which allows the agent or user to execute scripts on managed Azure Virtual Machines.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 22, 2026, 01:05 PM
Security Audit — agent-trust-hub — azure-compute