Grok Bot 抓网页省 token 实测：脚本比截图省 4 倍

Michael | AI + Agents @Michael_Fenech_原推发布于 2026.10.28

在 X 上打开原推

I ran an experiment and found a way to make Grok Bot more efficient: change how it reads websites. Before I give you the exact prompt to paste into your Chief of Staff, here’s what I tested. I asked Grok Bot to extract data from the same website three ways: by looking at the screen, scrolling and clicking; by using a script inside the browser; and by running a script without a browser. Here are the approximate tokens used from that practice-site test: 1: Computer use: 213K tokens. Screenshots, scrolling and clicking. 2: Browser plus script: 51K tokens. Open the page, then read its contents with code. 3: Plain script: no model tokens during execution. Fetch the pages directly, with model usage still needed around the run. The scripts also preserved full titles that computer use shortened. The takeaway: Grok Bot doesn’t always need to look at a website to read it. A script can collect the data without the Bot analysing screenshots and deciding where to click at every step. For repeatable website tasks, checking whether a script can do the job should be part of your setup. That’s an opportunity to save model work every time it runs. To add this process to your own Grok Bot setup, paste this prompt into your Chief of Staff: “Help my Bots use less model work when extracting information from websites, without sacrificing accuracy or completeness. Use this order, stopping at the first suitable method that works: 1: A connected app. Use an existing connector if it covers the task and returns the information we need. 2: A plain script. For public data or an authorised official data feed, fetch and process it directly. Return only the information needed for the task. Avoid opening a browser unnecessarily. 3: The browser, read by a script. If the page needs a browser, check whether supported browser scripting can read its contents accurately instead of repeatedly analysing screenshots. 4: Computer use. Use screenshots, scrolling and clicking for the remaining steps that need visual interaction or cannot be completed reliably by the methods above. Apply these rules: Check the result. Verify record counts, required fields and key details against the source. Don’t silently shorten, omit or invent information. Fail clearly. If a method fails or returns incomplete or suspicious data, report the problem before trying the next suitable method. Reuse successful scripts. Save tested scripts with instructions and validation checks so recurring jobs don’t require rebuilding them. Keep context lean. For repeat jobs, use a routine or a short, fresh chat with the necessary instructions and inputs. Avoid carrying unrelated conversation history into every run. Respect access limits. Use supported access methods. Don’t extract protected credentials or bypass sign-ins, CAPTCHAs or bot protection. Stop if human action is required. Report the method. Tell me what you used, why earlier options weren’t suitable, and any limitations. Report measured usage where available; label estimates. Propose this rule for the relevant Bots and routines, then wait for my approval before changing their instructions.” You don’t need to write the script. Give Grok Bot the rule for choosing when to use one.