test-changes

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs running LLM commands that inject raw repository diffs and specific TF files (e.g., via "$(git diff)" and "@locals.tf @variables.tf") into the model prompt, which will expose any API tokens or secrets present in those files to the model and may cause them to be echoed in output—creating a high exfiltration risk.