The Agent Skills Directory

Sensitive Data Protection: The deployment process includes explicit configuration to exclude sensitive files such as .env, .git, and virtual environment folders (.venv) when packaging application code. This practice effectively prevents the accidental exfiltration of secrets or credentials during the zip-based deployment process.
Shell Injection Prevention: The instructions include a specific security constraint mandating the use of the colon character (:) instead of the pipe character (|) for runtime specifications (e.g., PYTHON:3.14). This design choice proactively avoids potential shell interpretation issues and injection vulnerabilities across different operating systems.
Trusted Tooling and Official APIs: The skill relies exclusively on established Microsoft tools, including the Azure CLI (az) and the Azure Developer CLI (azd). All network operations and resource management tasks are conducted through these official, authenticated channels, ensuring high trust in the communication endpoints.
Robust and Idempotent Logic: The included helper scripts (retry-az-create.sh and retry-az-create.ps1) implement idempotent logic for resource creation. By performing a "show" check before attempting a "create" operation, the skill reduces configuration errors and prevents the creation of unintended or duplicate infrastructure components.
Managed Startup Configurations: The skill automates startup command generation for supported frameworks like FastAPI, ensuring that the application binds to the correct network interfaces (0.0.0.0). This reduces the risk of manual configuration errors that could lead to connectivity failures or misconfigured listener ports.

python-appservice-deploy