azure-postgres

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Potential surface for indirect prompt injection via variables in setup scripts. Ingestion points: CLI arguments (resource names, IDs, UPNs) in scripts/setup-user.sh, scripts/setup-group.sh, and scripts/setup-managed-identity.sh. Boundary markers: Absent. Capability inventory: Execution of Azure CLI (az) and PostgreSQL (psql) commands with administrative privileges. Sanitization: Absent; inputs are interpolated directly into shell and SQL strings.
  • [CREDENTIALS_UNSAFE] (SAFE): No hardcoded credentials, API keys, or sensitive files were detected. The skill correctly utilizes Azure's managed identity and access token mechanisms.
  • [COMMAND_EXECUTION] (SAFE): Use of shell scripts and SQL queries is appropriate for the documented database and cloud administration tasks.
  • [EXTERNAL_DOWNLOADS] (SAFE): Documentation references official and trusted packages from PyPI and npm registries.
  • [DATA_EXFILTRATION] (SAFE): No unauthorized network activity or data exfiltration patterns were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 03:34 AM