Skills
Navigate
skills/sickn33/agentic-awesome-skills/azure-vms/Gen Agent Trust Hub

azure-vms

Fail

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: HIGHCREDENTIALS_UNSAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill includes a hardcoded administrator password in a command-line example for creating a Windows VM.
  • Evidence: --admin-password 'S3cur3P@ssw0rd!' in SKILL.md.
  • [DATA_EXFILTRATION]: The Terraform configuration templates provided in the skill reference sensitive local file paths to read SSH public keys.
  • Evidence: public_key = file("~/.ssh/id_rsa.pub") in references/details.md.
  • [COMMAND_EXECUTION]: The skill documents the use of Azure CLI tools designed to execute arbitrary shell scripts on remote virtual machines.
  • Evidence: az vm run-command invoke in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external configuration data (cloud-init) which represents a surface for indirect prompt injection if the source file is manipulated by an attacker.
  • Ingestion points: cloud-init.yaml processed via the --custom-data flag in az vm create (SKILL.md).
  • Boundary markers: None present.
  • Capability inventory: Remote shell script execution (az vm run-command invoke) and virtual machine provisioning (az vm create).
  • Sanitization: None present.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Oct 22, 2026, 01:06 PM