Google adds Chrome 'Skills' and Fabula
Google introduced 'Skills' in Chrome — one‑click AI workflows for simple tasks like recipe swaps or document scans that users can save from chat history — and Google Research unveiled Fabula, an interactive writing tool co‑designed with writers to structure and refine stories. Both moves signal product pushes to embed short, task‑oriented AI features into browser and creative workflows. ( ).
Google is putting smaller, repeatable artificial intelligence tasks directly into Chrome and a writer-focused storytelling tool into Google Research’s demo lineup. (blog.google) On April 14, Google said Chrome users can now save a prompt from Gemini in Chrome chat history as a “Skill” and run it again with one click. Google’s examples include swapping recipe ingredients, comparing product specifications across tabs, and scanning long documents for key information. (blog.google) The feature works inside Gemini in Chrome’s side panel, where saved Skills can be called up by typing a forward slash or clicking the plus button. Google also launched a built-in library of prewritten Skills for common tasks and said some actions, including sending an email or adding a calendar event, still require confirmation. (blog.google) A browser “Skill” is essentially a saved prompt that can be reused on the page you are viewing, or across several tabs at once, instead of retyping the same instruction each time. Google said the system is built on the same security and privacy safeguards it uses for prompts in Gemini in Chrome. (blog.google) Fabula addresses a different problem: how writers use artificial intelligence without handing over the whole job to a chatbot. Google Research describes Fabula as “an interactive AI writing tool” that helps authors structure and refine stories, with the design shaped by 42 expert writers. (research.google) Google listed Fabula as a live demo at the Association for Computing Machinery Conference on Human Factors in Computing Systems, or CHI 2026, in Barcelona, scheduled for Wednesday, April 15. The conference page says the demo focuses on “convergent iteration,” a process for narrowing and improving ideas during drafting. (research.google) Both announcements build on Google’s broader push to turn Gemini from a chat window into a layer that sits inside other products. In February, Google said Gemini in Chrome was getting a redesigned side panel, deeper app integrations, and tools for handling multi-step work across the web. (blog.google) Chrome is a particularly important place for that push because Google already frames the browser as more than a page viewer. When Google expanded Gemini in Chrome in September 2025, it said the assistant could read the current page, compare information across tabs, and connect with Google apps such as Calendar, Maps, and YouTube. (9to5google.com) Fabula also fits a longer Google Research pattern of building creative-writing interfaces rather than only bigger language models. Earlier Google research projects, including Story Centaur, explored tools that let writers shape and recombine machine-generated text through custom interfaces. (research.google) Taken together, Skills and Fabula show Google splitting its artificial intelligence work into two tracks: quick browser actions for everyday tasks and guided creative tools for specialized users. In both cases, the interface is the product as much as the model behind it. (blog.google, research.google)
