The Agent Skills Directory

[SAFE]: The skill generates deployment configurations and CI/CD pipelines using predefined templates and user-supplied decisions. It explicitly implements security best practices by instructing the agent to avoid writing secrets, tokens, or credentials into generated files, recommending instead the use of platform-native secret management.
[SAFE]: External references (such as GitHub Actions, Terraform providers, and Docker base images) point to official repositories from trusted organizations including AWS, Google, and HashiCorp, or well-known registries like NPM and the Terraform Registry.
[SAFE]: The generation process utilizes multi-stage Docker builds and non-root users for container targets, and promotes secure authentication methods like OIDC (OpenID Connect) for AWS and GCP deployments.
[SAFE]: The workflow is strictly sequential and requires explicit user approval at multiple stages (decisions, planning, and generation), preventing autonomous or unintended file modifications.
[SAFE]: All network operations mentioned, such as web searches for tool versions or health check calls, are scoped to legitimate development tasks and target well-known, established services.

aidlc-deploy