html-parser-rule

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

• Third-party content exposure detected (high risk: 1.00). The skill instructs fetching arbitrary public URLs (e.g., "When user provides a URL to parse" and the curl command curl -L "{URL}" > /tmp/source.html) and then reading and parsing that HTML, which exposes the agent to untrusted third-party web content that could carry indirect prompt injection.