
What happened
The column shows that in Chrome's Gemini side panel, typing "/" in the prompt box brings up "Add skill", where a saved instruction can be named and reused later.
Why it matters
Readers who repeat the same requests while browsing can skip retyping long prompts, a small time saving the column frames as making everyday web browsing more convenient.
Summaries like this, every morning.
The piece is written as a beginner's guide rather than a product announcement, and its premise is that AI tools on phones and PCs have become ordinary enough that even people who have never really used them need hints. Within that frame, the interesting detail is how Google exposes repeated instructions: rather than a separate skill library, the entry point is the prompt box itself, where a slash summons the saved list.
The three examples are chosen to show what "a skill" is worth doing at all. The column concedes that simple requests like "summarize this" barely save effort over typing the prompt by hand, so its samples bundle several constraints — a fixed structure, a child-level vocabulary, a rule against taking sides — into one stored instruction. The shopping and gourmet-video examples go further, specifying output formats and telling the assistant to label guesses as possibilities rather than state them as fact.
Read together, the examples suggest the practical value lies less in the assistant's raw ability than in packaging a repeatable procedure once and reusing it, which the column presents as a way to make browsing more enjoyable rather than as a productivity claim.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
At a Simeji business meetup held at Google's Shibuya office on October 29, 2026, Simeji operating head Yu Furu…

The write-up describes giving large language models distinct pro, anti and neutral roles to debate policy prop…

AI Senatus publicly shows lawmaker-role AI agents debating policy, letting outsiders check how arguments unfol…

The guide explains AI agent ROI is calculated as (effect amount minus total cost) divided by total cost, with…

Vals tested AI models on simulated corporate networks; OpenAI's restricted Daybreak completed 81% of tasks, wh…

Startups including Handshake, Micro1, and Mercor are pitching companies on selling internal work files—documen…
