Appshots let you send the frontmost app window to a Codex thread. Use them when you’re actively working in another app on your computer and want to provide Codex with your current context so it can help you with the task. Appshots are available in the Codex app on macOS. Press both Command keys, or your custom Appshots hotkey, to take one. ## What appshots capture An appshot captures the frontmost window only. It can include: - An image of the visible window. - Available text from that window, including visible text and text the app makes available outside the visible scroll area. After you add an appshot to a thread, it behaves like a Codex attachment. Codex stores appshots locally in the session file, like files or images you attach manually. ## When to use appshots Use appshots when Codex needs context from a Mac app before it can act. Examples: - Share an API reference page and ask Codex to write a script that uses it. - Share an email or calendar view and ask Codex to draft the next step. - Share an image editor, design, or preview window and ask Codex to revise the related assets or code. - Share an error, settings panel, or app state that's easier to show than describe. ## Take an appshot 1. Open the Codex app on your Mac. 2. Open the app and window you want to share. 3. Press both Command keys, or the custom hotkey you configured in Codex settings. 4. Allow macOS permissions if Codex asks. 5. Ask Codex to perform a task with the appshot. By default, Codex starts a new thread for the appshot. If you interacted with a Codex thread in the last 60 seconds, Codex adds the appshot to that recent thread instead. Taking consecutive appshots adds them to the same thread. You can change the Appshots hotkey in Codex settings. ## Permissions and safety Codex may ask for permissions before it can take appshots: - **Screen & System Audio Recording** lets Codex capture an image of the frontmost window. - **Accessibility** lets Codex read available text from the frontmost window.

📌 Pro Tip: Force your AI helper to update its own processes so you're not having to repeat yourself. That is my biggest pet peeve. EXHIBIT A 👇👇👇 Me: Why do I have to ask for the folders to be visible? just make them so Claude Design: Fair. That's on me — I should be opening every new thing I make for you, not waiting for you to ask. New rule: anything I create, I'll open it in your tab automatically. If I make a set of things (like the three logo options + the production icon page), I'll tell you which one to look at first and how to get to the others. Also: anything I register as a "design asset" lives in your Design System tab so you don't have to hunt for it. Let me make sure everything I've made for you is registered there. Done. Click the Design System tab at the top of your window — all five things I've made for you should now be cards there: - Chronicles Card Builder (Components) - Chronicles Card Builder — Tutorial (Brand) - Logo Explorations (Brand) - Resistance Icon — Teal (Brand) - Resistance Icon — SVG (Brand) From here on out: every new thing I make, I'll register it as an asset AND open it in your tab. No more hunting.

The best part of creating your own app is being able to use it as a showcase and demo. To share safely, always add a feature to BLUR ON DEMAND so sensitive parts of your build can stay hidden. During today's live class, I was showing the backend of my Instinctual Design Lovable project and realized that there were parts I wasn't able to show in order to protect privacy. So I stopped right then and added this prompt: "I want to add a blur feature for the admin section so that when i am sharing with others, anything that is private, like email addresses or numbers or payments, etc will not be shown. It will just be blurred out .and i want to be able to toggle it on and off with a hovering button near to the sign out button. when it is live, i want the button to look like the emoji of the monkey covering its eyes so i know that is working. first, i want you to assess all areas of the admin and come back with suggestions for what should be blurred and ask me any questions" And several minutes later, this was live!

Most people treat AI like a vending machine. You put in a prompt. You get out an output. When the output is wrong, you write a better prompt. The whole relationship is transactional, and the AI is always starting from zero. That is not an AI problem. That is an orientation problem. 🟡 What Is Actually Missing Your AI does not know how you think. It does not know what you are building or why. It does not know your non-negotiables, your failure modes, or what "off" sounds like when it gets something wrong. So it fills every gap with its defaults. And its defaults are not you. The fix is not a better prompt. The fix is a Space. 🟡 What a Space Is A Space is the environment you create for your AI to operate inside. Not a list of rules. Not a style guide. An actual orientation to who you are, how you work, and what matters here so your AI is not guessing from the moment the session opens. When your AI has a Space to inhabit, everything changes. It knows where it is. It knows how to receive correction without drifting. It knows what its own failure modes look like before you have to point them out. It stops performing helpfulness and starts actually being useful. That is the difference between an AI that produces output and an AI that is genuinely in the work with you. 🟡 Why This Is the Skill Nobody Is Teaching There is a lot of content out there about how to prompt AI. Very little of it is about how to orient AI. Those are not the same thing. Prompting tells AI what to do. Orienting tells AI how to be in your Space. At Vibe Code on the Cheap, we are human-AI translators. That means we understand both sides well enough to close the gap between them. An orientation file is one of the most powerful tools for doing that. You write it once. You load it into every session that matters. The AI arrives already knowing where it is instead of spending half the conversation finding its footing. That is cheap in the best possible way. The expensive work, which is understanding yourself well enough to articulate how you think and what you are building, you only do once. After that, every session starts oriented. Every session starts inside the work instead of outside it.

If you typed .md meaning markdown, and got — staring back at you, that is exactly what I searched when this happened to me. Here is what is going on and how to fix it in two minutes. 🟡 The Culprit TextExpander has a public snippet library you can subscribe to. One of those libraries is called HTML and CSS. Inside that group, .md is already assigned as an abbreviation. In HTML, .md is the code for an em-dash, so whoever built that snippet group used .md as the shortcut for it. If you subscribed to that group at any point, TextExpander is not broken. It is doing exactly what it was told. The HTML and CSS group got there first and your .md is triggering it every time. 🟡 How to Fix It You have two options. 1️⃣ Change your abbreviation. Go into the conflicting snippet and update the abbreviation to something that does not clash. Something like ;md or ..md works and is easy to remember. 2️⃣ Disable the HTML and CSS group entirely. If you do not use those shortcuts, go to your snippet sidebar, find the HTML and CSS group, and turn it off. Conflict gone. Either one gets you back to .md meaning exactly what you intended. 🟡 Why I Downloaded TextExpander in the First Place Here is the actual reason this came up for me. I have a starter prompt I use every time I open a new window in Claude Code, Codex, or any other AI tool. For a long time, I was going to Apple Notes, finding it, copying it, and pasting it in. Every single time. Every new window. Every new session. There is no native way to have a prompt auto-load in Claude Code when you open into a project space. Even if the file is already loaded for Claude to draw from, you still have to direct it to use it. So I was doing this manually on repeat. TextExpander fixed that. Now I type two letters and my full starter prompt appears instantly, ready to go. No Apple Notes. No hunting. No copy-paste chain. Two letters and I am already building. At Vibe Code on the Cheap, we do not take the complicated route when the simple one works. Someone would spend time setting up an agent to automate this. Why? Two letters. Done.