terraform-validator

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

• Third-party content exposure detected (high risk: 0.90). SKILL.md's required "Context7 Fallback to WebSearch" step mandates performing WebSearch queries (e.g., "terraform-provider-{name} hashicorp documentation" / site:registry.terraform.io) when Context7 has no provider, and instructs the agent to read and include that provider documentation in the validation report, so the agent will fetch and interpret untrusted public web content that can materially influence remediation and next actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's wrapper scripts install remote Python packages at runtime (scripts/extract_tf_info_wrapper.sh runs pip install python-hcl2 — e.g. https://pypi.org/project/python-hcl2 — and scripts/install_checkov.sh runs pip install checkov — e.g. https://pypi.org/project/checkov), which fetch and install/execute remote code and are required for the validator to run.