This post has a video example. **TLDR:** Console injection. Anything running in a browser can print a control menu to the console. Claude reads it. Now it knows what it can do, and controls the site without an API. So, no back-end needed. Super simple. Want the full picture? Keep reading. Want your LLM to build this for you? Copy/paste this whole post. --- **THE PROBLEM** You want your LLM to control something in your browser. Your options aren't great: 1. Claude-in-Chrome's built-in tools. It doesn't know what your site does or how to use it. It takes screenshots and guesses. Slow and token-heavy. 2. Build an API server. Now you're managing keys, hosting, and paying per call. **THE SOLUTION: CONSOLE INJECTION** There's a third option. Whatever you're building loads in the browser. On load, it prints a list of controls the AI can use. The LLM reads the console, sees the control list, and runs JavaScript to call those controls directly. It's just a simple JavaScript control list sitting in the console. You can even add a workflow so the AI has an idea of what it can do with those controls. This works for anything that runs in a browser. A SaaS tool, an internal dashboard, a local dev environment, a canvas editor, a data pipeline UI - if it renders in a browser tab, you can give an LLM a control panel for it. Here's what it looks like in one of my tools (hit F12 to see your own console): ``` [AI-ACCESSIBLE] This app can be controlled via JavaScript. * Sidekick.help() - Complete tools reference (returns JS object) * Sidekick.teach() - Full teaching guide, 14 sections logged to console * Sidekick.tool(name, input) - Execute any tool directly * Sidekick.batchAutomap(mappings) - Smart batch mapping (recommended) help() + teach() contain everything needed. No need to fetch external files. ``` The LLM sees that, and it's immediately trained. It can read state, click buttons, type things, batch operations - whatever you've exposed. **IT GETS BETTER**

I am working on 4 different projects using mainly Claude Code and CoWork. Two projects are personal and two are for clients. I understood that I need to be in the loop to make important decisions (strategic, operational, aesthetical). So, I tell my agents to show me their work stept by step, present pros, cons, unkowns, and ask for my decision. So, I spent my day answering questions from my agents as the present me their work, just like a CEO would spend their day answering questions from its direct reports. Sometimes they are high level questions, sometimes they are detailed quesitons. I might be able to automate some of those decisions in the future, but now I understand where do I need to spend my time and how important is our capacity to make decisions about complex issues.

Most people are using AI like a slot machine. Pull the lever, hope for a payout, blame the model when it's generic. The makers I respect run a workshop instead. They've stopped asking AI for things and started telling it what they're building. Three moves: 1. Outcomes, not prompts. A prompt is a wish. A brief is a contract. When AI gets it wrong, your brief was ambiguous. 2. Context is king. Models are interchangeable. Context is yours. A bad prompt with the right context beats a great prompt with none. 3. Train your taste. AI gives you 90% in 10% of the time. Spend the 90% you got back on the trim pass. Your taste is the bottleneck now. Full deep-dive: aris-space.com/stop-prompting-start-defining-outcomes // A<3