The Agent Skills Directory

[Command Execution]: The skill executes Azure CLI commands and application-specific migration scripts (e.g., Django, Alembic, or Prisma) to provision resources and initialize the application environment. This behavior is mediated by a detailed list of blocked patterns to prevent the execution of potentially harmful commands like recursive deletion or privilege escalation.\n- [Secret Management]: The skill uses Azure Key Vault and secure parameters to handle sensitive credentials. It employs a local session file to persist secrets across tool calls within a specific session context, which is standard for local development workflows. It explicitly instructs the agent to avoid hardcoding secrets in infrastructure templates or logging them in chat history.\n- [External Package Installation]: The skill uses package managers to install official tools like the Azure Static Web Apps CLI when needed. These downloads target official registries and are limited to tools required for the deployment process.\n- [Workspace Analysis and Data Ingestion]: The skill analyzes repository configuration files (such as .env.example and docker-compose.yml) to derive infrastructure requirements. This ingestion is managed through structured planning and validation phases, including a conformance gate that checks generated code against the original plan for semantic errors.

azure-app-onboard