fabric-builder

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow reads outsider-authored free text from GitHub module files (e.g., README/variables/outputs/schema.json) by fetching them via scripts/fabric.py (api_get/fetch_url -> fetch(RAW_URL/...)-> raw_data.decode(...)), where those files are authored by GitHub users/committers to the referenced repository.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime helper scripts fetch module content from GitHub (e.g., f"https://api.github.com/repos/{REPO}" and f"https://raw.githubusercontent.com/{REPO}/master") and the SKILL explicitly requires using those fetched READMEs/variables to drive Terraform-generation instructions, so external content from https://raw.githubusercontent.com/GoogleCloudPlatform/cloud-foundation-fabric/master (via the script) directly controls agent behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs running the helper script with a --no-ssl-verify flag to bypass SSL certificate verification on the host, which encourages bypassing a security mechanism.