azure-security-lab-terraform
Warn
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions clone a repository from a third-party GitHub account (Jamonygr/Azure-Security-Lab) that is not part of a verified or trusted organization.
- [REMOTE_CODE_EXECUTION]: The downloaded Terraform code is executed using terraform apply, which can run arbitrary logic through providers or provisioners.
- [COMMAND_EXECUTION]: The skill uses terraform, az CLI, and git commands to manage cloud infrastructure and authentication.
- [INDIRECT_PROMPT_INJECTION]: The skill processes Azure Firewall logs which ingest external network traffic data. This creates a surface where malicious strings in traffic (like DNS names) could be interpreted as instructions by the agent. Evidence Chain: Ingestion point is AzureDiagnostics logs in SKILL.md. Capability inventory includes cloud resource management and credential access. Boundary markers are not present. Sanitization is not present.
Audit Metadata