Web Reader

This tool fetches the HTML content at a specified URL and returns a cleaned, readable Markdown representation with boilerplate, ads, and navigation removed, optionally retaining image links. It accepts two inputs: url (required string, the target web page URL) and include_images (optional boolean, when true image links are kept in the Markdown; otherwise they are removed). The worker must transform these inputs and any execution results into a final JSON response that strictly validates against the system-provided structured output schema for this tool, with no extra fields and correct types.