The Agent Skills Directory

[SAFE]: The skill provides architecture patterns for AWS infrastructure (CDK, Terraform, SAM, CloudFormation) that incorporate security features like cdk-nag and remote state locking.
[SAFE]: External dependencies are restricted to official AWS MCP tools for documentation lookup and resource validation, which are recognized vendor-provided resources.
[SAFE]: There are no indicators of malicious intent, obfuscation, or unauthorized data access. The skill functions as a template-driven code generator.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface typical for code generation tools that process user descriptions and external documentation.
Ingestion points: User input through $ARGUMENTS[1] and data fetched via awsknowledge MCP tools in SKILL.md.
Boundary markers: No explicit delimiters are implemented to encapsulate untrusted data within the generation process.
Capability inventory: The skill is designed to generate complex infrastructure files and deployment pipelines.
Sanitization: Input processing relies on standard model behavior without additional validation layers.

iac-scaffold