security-scan

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly runs curl checks against public URLs (e.g., "curl -I https://example.com/.git/config", ".env", "var/log/system.log") as part of its required file-exposure scanning, which means the agent fetches and interprets content from open/public sites that could be untrusted and influence scanning decisions.