aws-ecs-fargate
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the management of cloud infrastructure by consuming external configuration data, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill processes JSON task definitions (e.g., 'task-definition.json'), environment variables, and Terraform HCL files.
- Boundary markers: No explicit delimiters or warnings to disregard instructions within the data (e.g., within container log streams or task definition metadata) are included in the provided snippets.
- Capability inventory: The skill has extensive capabilities, including creating clusters, pushing images, registering tasks, and executing shell commands within running containers via 'aws ecs execute-command'.
- Sanitization: The skill relies on the validation mechanisms of the underlying AWS and Terraform binaries but does not explicitly sanitize data for natural language instructions.
Audit Metadata