The Agent Skills Directory

[Shell Command Execution]: The skill relies on standard Azure CLI (az) and Azure Developer CLI (azd) commands to provision infrastructure and deploy application code. These operations are core to the skill's functionality for managing Azure resources.
[Secure Data Handling]: During the packaging phase, the skill explicitly instructs the exclusion of sensitive files and directories such as .env, .git, and virtual environment folders (.venv). This is a recommended practice to prevent the accidental exposure of credentials or unnecessary metadata.
[Sanitized Resource Generation]: The inclusion of helper scripts (generate-app-name.sh and generate-app-name.ps1) demonstrates a proactive approach to security. These scripts use regular expressions to sanitize folder names into valid Azure resource slugs, effectively preventing command injection or naming conflicts during resource creation.
[Transient Error Management]: The skill uses dedicated scripts (retry-az-create.sh/ps1) to handle transient cloud infrastructure errors. While these scripts use dynamic execution (such as eval or PowerShell's Invoke-Expression patterns) to perform retries, they are used exclusively with well-defined Azure CLI templates provided within the skill's instructions.
[Framework Detection Logic]: The skill includes logic to detect the application framework (Flask, Django, or FastAPI) to suggest appropriate startup commands. This detection is used to improve the deployment experience and ensure the application starts correctly in the cloud environment.

python-appservice-deploy