The Agent Skills Directory

[External Configuration Inspection]: The skill uses curl to retrieve infrastructure-as-code files from official Microsoft Azure sample repositories. This activity is used to analyze and verify security settings (such as authentication methods and firewall rules) against recommended baselines before deployment.
[Environment Variable Integration]: The skill utilizes eval to load environment variables generated by the Azure Developer CLI tool (azd env get-values). This is a standard pattern for synchronizing infrastructure outputs with shell scripts during deployment hooks.
[Security Remediation Guidance]: A primary focus of the skill is identifying and correcting security configurations in common templates. It recommends replacing password-based authentication with Entra ID managed identities and narrowing overly permissive firewall rules to follow the principle of least privilege.
[Project Data Ingestion]: The skill instructs the agent to process local project files to configure the deployment environment, which presents a surface for processing configuration data.
Ingestion points: The skill reads azure.yaml, ./infra directory contents (Bicep files), and azd environment values.
Boundary markers: It uses specific CLI commands (azd env get-values, grep) to isolate and interpret relevant infrastructure data.
Capability inventory: The skill uses shell commands via azd, az, sqlcmd, and curl for deployment and verification.
Sanitization: It emphasizes reading values directly from the azd environment rather than using hardcoded or guessed resource names, which reduces the risk of configuration errors.

deploy-app-to-azure