resilience-management-operations
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution Fallback: The instructions allow for the use of standard command-line interfaces and direct service calls if specific tools are unavailable. This is a common feature for comprehensive resource management and is regulated by a requirement for explicit user authorization before execution.
- Data Interaction Surface: The skill ingests and processes user-provided strings (identifiers, descriptions) and responses from Azure services. This represents a typical interaction surface for management tasks, which is mitigated by structured inputs and platform-level safety guardrails.
- Ingestion points: User inputs and service response data processed in SKILL.md and references/workflows.md.
- Boundary markers: The skill mandates user confirmation for destructive actions and fallback operations.
- Capability inventory: Operations include calls to registered MCP tools, with optional fallbacks to system command-line tools.
- Sanitization: The skill relies on standard Azure Resource Manager validation and MCP schema enforcement.
- Development Utility Execution: Reference documentation describes the use of local scripts and specialized binaries for building and testing the toolset. These are intended for development environment setup and are distinct from the skill's operational runtime.
Audit Metadata