Paste a URL and receive a concise summary of the page's key points. The crawler fetches the visible text, strips nav/ads, and passes it to an LLM for condensation. Good for quickly triaging long articles, blog posts, and documentation pages before committing time to a deep read. Maximum context window (input and output combined): about 1,048,576 tokens. Maximum output per reply: about 65,536 tokens (model reasoning, if any, counts toward this limit). Output content formats: Text This module does not support continuous conversations. AI can make mistakes. Please verify important information.