azure-validate
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- Prompt Injection (LOW): The skill uses authoritative override markers such as 'AUTHORITATIVE GUIDANCE' and 'supersedes prior training' in its instructions. These are common prompt injection patterns used here to maintain agent focus on the validation workflow.
- Indirect Prompt Injection (LOW): The skill processes untrusted content from .azure/plan.md without explicit delimiters or sanitization. Ingestion points: .azure/plan.md. Boundary markers: Absent. Capability inventory: Execution of cloud provider CLI commands (az, azd, terraform). Sanitization: None detected.
- External Downloads (LOW): The skill references and triggers the installation of common Azure development tools (azd, az CLI) via specialized MCP tools. These are considered trusted dependencies in this context.
- Command Execution (LOW): The skill's primary function involves running infrastructure-as-code validation commands. Safety is partially addressed through global rules requiring user confirmation for destructive actions.
Audit Metadata