api-gateway
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied requirements to generate specialized AWS Infrastructure as Code templates (SAM, CloudFormation, CDK) and VTL mapping templates. This creates an inherent attack surface for indirect prompt injection where malicious input could influence the generated configuration.
- Ingestion points: User queries and requirement specifications gathered through the workflow described in
references/requirements-gathering.md. - Boundary markers: The skill does not explicitly define delimiters for user-provided data when interpolating it into generated templates.
- Capability inventory: The skill facilitates the creation of resources with significant capabilities, including network access (VPC Links, Private APIs) and interaction with critical AWS data services (DynamoDB, SQS, EventBridge).
- Sanitization: The skill demonstrates security best practices in its reference templates (such as in
references/service-integrations.md), such as using$util.escapeJavaScript()and$util.urlEncode()to sanitize data within VTL mappings. - [SAFE]: The skill represents legitimate technical documentation and development tools for Amazon API Gateway. All external references target official AWS repositories and well-known documentation sources. The instructions emphasize security controls like IAM authorization, WAF integration, and mTLS, reinforcing a defensive security posture.
Audit Metadata