terraform-ci

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds secrets into CLI arguments (e.g., -var="db_password=${{ secrets.DB_PASSWORD }}") and reads/prints the terraform plan file to a PR comment — both patterns can cause secret values to be included verbatim and exfiltrated.

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Infracost Cost Gate" runnable CI step in SKILL.md explicitly fetches and executes a public script via curl from https://raw.githubusercontent.com/infracost/infracost/master/scripts/install.sh (public GitHub), which ingests untrusted third‑party content into the CI run and can materially influence gating/automation behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The workflow runs a runtime install step that executes remote code via "curl -fsSL https://raw.githubusercontent.com/infracost/infracost/master/scripts/install.sh | sh", which downloads and immediately executes a script the CI relies on to run Infracost.