aws-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs discovery of /.aws/credentials and to include "Evidence: " from discovered files, which can require the LLM to read and output AWS access keys or secrets verbatim.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The required workflow ingests repository-discovered Terraform/CloudFormation/IAM/policy/config file contents (outsider-authored if the repo includes third-party text) as readable prose into the LLM context for evaluation, creating an indirect prompt-injection risk from those files’ string values/comments/descriptions.