内容来源：https://www.wired.com/story/how-to-use-google-chrome-ai-powered-skills/

内容总结：

谷歌浏览器近日推出名为“技能”（Skills）的生成式AI新功能，用户可通过键盘快捷键快速调用预设AI指令，进一步强化其产品生态的AI整合布局。

该功能依托于侧边栏的Gemini助手实现，用户既可自定义“技能”，也可从官方提供的超过50种预设模板中选择，涵盖视频摘要、食谱优化、招聘信息评估等多样化场景。以“蛋白质最大化”技能为例，系统可自动分析网页食谱成分，在保持风味的前提下推荐高蛋白替代方案，并生成详细营养数据。

用户点击浏览器右上角“询问Gemini”星形图标即可启用该功能。在提示框输入斜杠后选择具体技能，Gemini将根据授权对当前浏览内容进行智能分析。若无需此功能，用户可在设置中关闭AI创新选项以隐藏相关按钮。

行业观察显示，此类浏览器AI工具虽在初期可能存在稳定性问题，但已吸引注重效率的用户群体。值得关注的是，挪威Opera Neon浏览器此前已推出类似“卡片”（Cards）功能，提供旅行规划、影夜安排等热门模板，反映出浏览器赛道正普遍向可复用AI指令方向演进。

此次更新是谷歌今年推动浏览器AI化的重要举措。此前该公司已将Gemini助手深度集成至侧边栏，并尝试让AI代理用户执行网页操作。随着各厂商持续加码智能功能，浏览器正从信息工具逐步转向智能工作平台。

中文翻译：

谷歌浏览器近日又添一项生成式AI功能：技能。这是一组可通过键盘快捷键在浏览器中运行的AI指令模板，为谷歌全线软件注入AI工具的清单再添新成员。

用户可通过Chrome浏览器调用谷歌聊天机器人Gemini创建专属技能，亦可直接选用谷歌随功能发布的预设技能库。技能库包含50余种预设模板，涵盖YouTube视频摘要、通过食谱替换实现蛋白质摄入最大化、职位招聘信息评估等多种指令场景。

若想体验该功能，请点击屏幕右上角的"询问Gemini"星形图标，打开浏览器侧边栏的Gemini界面。在指令框中输入斜杠符号，即可选择要运行的技能。选定后，Gemini将根据技能设定的参数范围，分析您已分享的浏览器标签页信息。

以谷歌官方示例"蛋白质优化器"技能为例，完整指令如下：
分析当前网页食谱，识别所有食材并估算其蛋白质含量。在保持原食谱风味的前提下，提出提升总蛋白质含量的食材替换或增补方案。输出改良版食谱，需列明每种食材的蛋白质含量及每份餐食的总蛋白质值。

基于我在不同浏览器测试生成式AI功能的经验，这类工具上线初期出现运行瑕疵不足为奇，预计未来数月将逐步完善。可以预见，这种旨在简化工作流程、减少点击次数的浏览器工具极易受到效率达人的青睐。即便如此，多数Chrome用户日常上网时恐怕根本不会意识到这项AI功能的存在。

对此功能不感兴趣但仍需使用Chrome的用户，可通过设置菜单进入"AI创新"选项卡，在"Chrome中的Gemini"版块关闭顶部开关，即可移除浏览器顶部的询问按钮。

今年初谷歌加速推进Chrome浏览器的AI化改造，在侧边栏集成Gemini助手，将其定位为常驻屏幕右侧的智能伴侣，随时解答网页内容相关问题。该公司还尝试让生成式AI接管浏览器操作，实现代用户点击浏览网页的功能。

在提升AI指令复用性方面，谷歌并非孤军奋战。挪威小众浏览器Opera Neon推出的"卡片"功能与之异曲同工，用户既可复用自定义指令，也能选用预设模板库。该平台最受欢迎的指令包括任务优先级排序、电影之夜策划、特价行程规划等工具。

英文来源：

Google Chrome just got another generative AI feature: Skills. Skills are repeatable AI prompts you can run in Chrome with a keyboard shortcut. Add it to the laundry list of AI tools Google has been injecting into all of its software.
You can set up your own Skill using Gemini, Google’s chatbot, through the Chrome browser, or you can choose from the premade Skills Google released alongside this feature. The more than 50 presets in the Skills library cover a range of prompts that instruct Gemini to summarize YouTube videos, maximize your protein intake via recipe substitutions, or evaluate job listings.
If you want to try out Skills, open up the Gemini in Chrome sidebar by clicking on the “Ask Gemini” sparkle icon in the upper-right corner of the screen. Then, type a forward slash in the prompt box to pick which Skill you would like to run. After you select one, Gemini analyzes the information from the browser tabs you’ve shared, within the parameters of the details laid out in the Skill.
Here’s the full prompt from Google’s example of a “Protein Maximizer” Skill to show how these can be used to guide Gemini in Chrome:
Analyze the recipe on the current webpage, identify all ingredients, and estimate their protein content. Suggest substitutions or additions to maximize the overall protein content of the recipe, while maintaining the integrity of the original recipe's flavor profile. Output the revised recipe with protein content listed for each ingredient and the total protein per serving.
From my experience testing generative AI features in different browsers, I wouldn’t be surprised if these tools were a bit glitchy at launch and gradually improved over the next few months. It’s also easy to imagine this kind of browser tool catching on with productivity nerds looking to streamline workflows and save clicks. Even so, Skills seems like the kind of AI feature most Chrome users probably won’t even realize is an option as they’re browsing the web.
Users who aren’t interested in this feature but still want to use Chrome have the option of removing the Ask Gemini button by going into their Settings and opening the AI Innovations tab. Then, open the Gemini in Chrome section and make sure that the top toggle on that page is turned off. When this setting is toggled off, the Ask Gemini button disappears from the top of the Chrome browser.
Google’s rework of its Chrome browser for the AI era intensified earlier this year with the addition of Gemini in the Chrome sidebar, pitched as an always-present assistant sitting on the right side of your screen, ready to answer questions about what you’re seeing on the web. The company has also experimented with how generative AI can take control of Chrome to click and browse the web on a user’s behalf.
Google is not alone in its attempts to make AI prompts more easily repeatable for users. The Opera Neon browser, a smaller Chrome competitor based in Norway, has a similar tool, called Cards, where users can reuse their own prompts or pick from a preset library. Some of the most popular options available on Opera Neon include tools for prioritizing tasks, planning movie nights, and scheduling cheap travel.