The Agent Skills Directory

Standard Command-Line Tooling: The skill utilizes official AWS CLI commands and the Amplify ampx utility to manage cloud infrastructure, deploy backend resources, and configure environment secrets. These operations are essential for the skill's functionality in a development and deployment context.
Trusted External Dependencies: The instructions reference official AWS Amplify libraries and project templates hosted on GitHub. These resources are provided by a well-known service and are used to facilitate standard project scaffolding and integration.
Indirect Prompt Injection Surface: The skill acts as a developer assistant that translates natural language requirements into TypeScript-based infrastructure definitions and application code. While this introduces a surface where malformed user input could theoretically influence the generated cloud resources, the skill mitigates this by emphasizing secure patterns like owner-based authorization and the use of the AWS Amplify secret management system for sensitive data.
Ingestion points: User prompts describing application features, data models, and framework requirements in SKILL.md and references/scaffolding.md.
Boundary markers: The skill relies on the underlying LLM's context handling and does not implement explicit delimiters for user-provided data within its instructions.
Capability inventory: File system access for project scaffolding, shell command execution for deployment (npx ampx), and management of AWS cloud resources through the aws CLI.
Sanitization: Promotes the use of secret() for credentials and environment variables, and provides guidance on implementing restrictive authorization rules (allow.owner()) for data models.

aws-amplify