Blog

AI pills,
for real work.

Practical guides on AI, websites and automation, grouped by what you want to do. Base if you are starting out, Pro if you already know some tech.

Newsletter

AI that actually works, in your inbox.

New pills, the tools we use and what we learn on real projects. Few emails, only when there is something useful.

No spam. Unsubscribe in one click.

Agents and automation

Repetitive work handed to AI agents, with real examples.

Agents and automation · 15 Nov 2026BaseCode Mode: Claude now does things in one goClaude Code now does several steps in one go, not one at a time. What changes, when it helps and when it doesn't. Still experimental.Agents and automation · 13 Nov 2026BaseClaude Meetings and Claude Sheets: Claude Joins Your Calls and Your Spreadsheets. What Changes for a Small BusinessAnthropic introduced a small work suite inside Claude: Meetings for video calls, Sheets for spreadsheets, Whiteboards for shared boards and Claude Data for company data. What each one does, who it is for and what it costs during the beta.Agents and automation · 13 Nov 2026BaseBusinesses in ChatGPT: OpenAI Opens Business Profiles. What a Small Company Can Do TodayOpenAI launched Businesses in ChatGPT in beta: companies claim their profile, update their information and see how people find them. Who can do it, how it works and what it means outside the US.Agents and automation · 12 Nov 2026Basen8n Just Launched Nod: A Free Personal Agent That Asks Before It ActsNod is the new personal agent from the team behind n8n: open source, free, runs on your own computer and asks before it acts for you. What it is and who it is for.Agents and automation · 11 Nov 2026BaseGrokBot: 3 Things You Might Not Know It Can DoGrokBot can receive email at a dedicated address, use websites without a connector and run cloud routines while your laptop is closed. How to start.Agents and automation · 10 Nov 2026BaseGrok Bot Hit My Usage Limit: How I Keep a Studio Running AnywayWhen Grok Bot runs out of usage, routines stop. Four simple rules to decide what runs first, what waits, and what to do when a run fails.Agents and automation · 8 Nov 2026ProPhoneUse: An AI Agent That Uses Your Real iPhoneCan an AI agent use a real iPhone with no app and no jailbreak? PhoneUse does it with AirPlay and Bluetooth. What it is, how it works, how to try it.Agents and automation · 8 Nov 2026BaseRun Grok Bot Like a Studio: 38 Tips for Your Bot TeamHow do you run several Grok Bot bots without losing control? 38 practical tips in three levels, with prompts to copy and a 31-page PDF guide.Agents and automation · 8 Nov 2026BaseMake Claude Code Use Less: 7 Habits and a Free PluginOne Claude Code chat for everything, and the tokens run out? Seven habits to spend less, and Quack, the free plugin that reminds you of them on its own.Agents and automation · 8 Nov 2026BaseMake Grok Bot Use Less: 9 Free Tools to InstallDoes your Grok Bot usage run out by midweek? Nine free tools make the Bot read only the useful part of web pages, PDFs and spreadsheets.Agents and automation · 8 Nov 2026ProDuplicate Webhooks: Stop the Same Order Appearing Twice in Your CRMWebhooks can arrive more than once. Design retries, deduplication and CRM writes so one payment does not create duplicate orders or notifications.Agents and automation · 8 Nov 2026ProAI Agent Memory: 5 Rules to Stop Repeating MistakesDoes your AI agent keep making the same mistakes? Five rules to give it a memory in markdown: what it reads when a session starts and writes at the end.Agents and automation · 8 Nov 2026ProParallel AI Agents: 5 Rules to Avoid the ChaosHow do you get several AI agents to work together without clashing? Five rules tested on real projects: split tasks, small teams, roles and routing.Agents and automation · 15 Sep 2026BaseAI Audit for Small Businesses: 5 Steps Before the ModelHow do you know if a business process is ready for AI? Five steps for a small-business AI audit: data, permissions, checks, a simple score, a first test.Agents and automation · 11 Sep 2026ProAI Agent Workspace: 5 Parts That Make Agents ReliableWhy does an AI agent go wrong even with a strong model? The five parts of a reliable workspace: brief, permissions, checkpoints, flow and metrics.Agents and automation · 10 Apr 2026ProAI Skills for Business Procedures: 7 Steps to 4 MinutesHow do you turn a repetitive internal procedure into a skill an AI agent can run? A real onboarding that went from seven manual steps to four minutes.