The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests user-provided CDK infrastructure code and Lambda handler source files, which presents a surface for indirect prompt injection. * Ingestion points: The skill processes user project files, such as TypeScript or Python stack definitions and Lambda entry point files (e.g., lambda/handler.ts). * Boundary markers: The instructions do not define boundary markers or safety delimiters to mitigate the risk of the agent following instructions embedded in code comments or metadata. * Capability inventory: The skill utilizes CloudFormation synthesis commands (cdk synth), executes local validation scripts (./scripts/validate-stack.sh), and installs development dependencies via package managers. * Sanitization: There is no evidence of input validation or sanitization for ingested files before they are processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill references the installation of tools and libraries from well-known repositories. * It recommends installing awslabs.aws-iac-mcp-server from the official AWS Labs organization using the uvx tool. * It suggests installing the cdk-nag package and other AWS CDK libraries from the npm registry for validation and development purposes.

aws-cdk-development