AI Skill of the Day Digest — September 2026 (Part 1)

Every day, The Neuron teaches readers one practical AI skill they can use immediately. This first September collection includes every published AI Skill of the Day from September 1 through September 15, 2026.

Skim the headings, grab the prompts that fit your work, and return when you need a better way to use Claude, ChatGPT, Gemini, Codex, or an AI agent.

How to use this digest
🎓 September 1
🎓 September 2
🎓 September 3
🎓 September 4
🎓 September 6
🎓 September 7
🎓 September 8
🎓 September 9
🎓 September 10
🎓 September 11
🎓 September 13
🎓 September 14
🎓 September 15
Keep learning

How to use this digest

Skimming? Each entry starts with the practical outcome.
Trying one? Copy the included prompt or workflow and replace the bracketed details.
Catching up? The skills are ordered by their original newsletter date.

🎓 September 1

Make AI report only what changed

Ever ask AI for a project update and get the entire history back? If your team uses Asana, ChatGPT can now read tasks, comments, owners, activity notes, and due dates through the Asana app.

That means you can ask for only what changed: new blockers, overdue work, owner changes, deadline shifts, and decisions somebody needs to make.

Connect Asana from ChatGPT’s app directory or Settings > Apps.

Give it a time window, then name the changes you care about.

Ask it to cite the task or comment behind each update. Review everything before letting AI change anything.

Prompt:

Review my Asana projects using recent task comments, activity notes, assignees, and due dates. Report only what changed since [DATE]: new blockers, newly overdue work, owner changes, due-date changes, and decisions someone needs to make. For each item, name the task and the source comment or activity note that supports it. Do not create or edit anything.

🎓 September 2

Make Fable 5.1 cheaper, faster, and less annoying

Anthropic published a surprisingly useful Fable 5.1 prompting guide. The biggest lesson: old Claude prompts mostly still work, but long-running agents need a few new habits.

Re-test effort from scratch. Start at the default high, then sweep low, medium, xhigh, and max on your own work. Anthropic says medium can roughly match Fable 5 at lower cost.

Cache repeated context. Fable 5.1 cache reads cost $0.25 per million tokens, down 75%, so repeated instructions, code, and documents are much cheaper to reread.

Keep conversation history append-only. Pass Claude’s earlier turns back exactly as returned, including thinking blocks. Rewriting older messages can invalidate both thinking and cached context.

Use mid-conversation system messages instead of rewriting the original system prompt when instructions change. Anthropic’s beta per-message effort control can also change effort without breaking the cached prefix.

Ask for progress updates on long jobs. Fable 5.1 narrates less between tool calls, so enable progress updates or explicitly request short status notes.

Batch independent tool calls. Anthropic says Fable 5.1 can otherwise fall into one-tool-call-per-turn loops that make agents slower than necessary.

Tell it to finish the whole assignment. Anthropic found the model may stop to ask permission for work you already requested unless autonomy is explicit.

Keep scope tight. Tell Claude not to “helpfully” fix nearby bugs, expand the feature, or add extra tests unless your task requires them.

At low effort, tell it when to search. Fable 5.1 is more willing to answer from memory there, so require verification for fast-moving names, tools, and models.

Ask for surgical file edits. Anthropic specifically recommends telling Fable 5.1 not to rewrite a whole file when a small patch gets the same result.

If the writing gets dense, tell it to “remove all mannered prose.” Anthropic says Fable 5.1 can write longer sentences with fewer paragraph breaks than Fable 5.

For false-positive coding refusals, avoid “does this compile?” phrasing, give context for obscure languages, and keep base64 blobs out of tool output when possible.

Give vision agents crop and zoom tools. Dense charts and images improve when Claude can inspect small regions instead of staring at one full-resolution frame.

Leave extra token room at xhigh and max. Fable 5.1 can spend much longer thinking before a large deliverable, so a tight max_tokens cap can cut off the answer.

One caveat for API builders: changing the ordinary request-level output_config.effort can invalidate cached message blocks. Use Anthropic’s per-message effort path if cache continuity matters.

🎓 September 3

Turn Your AI Tabs Into a Status Board

So '‘Computer Use’is the mode where OpenAI’s Codex or Claude Code can actually operate your computer for you, instead of only giving instructions or writing code. It can see what’s on screen, click buttons, type into websites, and work inside supported apps.

In the ChatGPT desktop app, open Codex, give it a task that requires a browser or app, then approve Computer Use when prompted. For browser work, you can also open Codex’s built-in browser from the toolbar and let it work across tabs.

Then, you can use this trick from Dan Shipper of Every; it’s a little developer-y, but the trick is simple: make the title of every AI task act like a status light. If you have several ChatGPT, Claude, or Codex jobs going at once, you should be able to tell what each one is doing without opening every tab.

Dan’s convention is: [optional emoji, like 🖥️] [status] [project] [task].

⚠️ means the job is still running or needs attention.

✅ means it is completely done.

🖥️ only appears while that specific agent is actually controlling a browser or desktop app, so you can instantly see which session can currently click or type on your computer.

You don’t need Codex computer control to steal this. Use the same ⚠️/✅ + project-emoji system for any long-running AI work, and update the title yourself or ask the agent to keep it current when your tool supports that.

For this task, keep the session title updated using:
[optional 🖥️] [status emoji] [project emoji] [task title]

Rules:
- ⚠️ = ongoing work or unresolved items.
- ✅ = fully complete.
- Use exactly one 🖥️ at the start only while you are actively controlling a browser or native app. Remove it when that phase ends.
- Shell commands, API calls, and ordinary web searches do not count as computer use.
- Preserve the project emoji and task title. Do not rename unrelated tasks.

Example: 🖥️ ⚠️ 🎬 Edit launch video while the agent is clicking around your editor, then ✅🎬 Edit launch video when the job is finished. Your tab bar becomes a tiny live dashboard instead of a row of mystery chats.

This system tells you which AI task currently has permission to click and type on your computer. And even if you never use Codex Computer Use, steal the simpler trick: label long-running ChatGPT or Claude tasks with ⚠️ while they’re still working, ✅ when they’re done, and a project emoji so your tabs become a tiny status board.

🎓 September 4

Give agents the “why” first

Agents can follow every instruction you give them and still build the wrong thing because they never understood why the project existed.

Anthropic's AI-native software playbook, highlighted in Rob Shocks' walkthrough, starts with a tiny file called intent.md. Think of it as the note you would leave for a smart coworker taking over the project tomorrow.

Describe the outcome and who it helps. Add the non-negotiable constraints and a concrete definition of done.

Let the agent interview you until the fuzzy parts are gone.

Save the answers as intent.md, then use that file to create the specification and plan.

The same habit works outside coding. Before a long research, writing, or analysis job, give the agent a one-page brief it can keep returning to.

🎓 September 6

Re-tune your instructions for GPT-6 Astra

GPT-6 Astra follows long instructions more closely, so rules written to compensate for older models can now create some friction. Eric Provencher and Angel Brodin recommend auditing the setup around Astra before adding more prompts.

Audit old rules first. Check AGENTS.md and Skills for instructions that force extra reading, approval, testing, or clarification.

Define “done,” not every step. State what should be implemented, inspected, fixed, and verified, then let Astra choose the route.

Scale testing to risk. Astra tends to test thoroughly, so limit broad “testing” on tiny, reversible changes instead of making every fix a full test run.

Shrink Skills into routers. Keep descriptions short and load detailed docs, examples, or scripts only when the task actually needs them.

The goal is to remove stale scaffolding while keeping the boundaries that actually matter to let your agents go off and cook.

🎓 September 7

Run Your Own Agent Orchestration With Just Browser Tabs

The best AI setups in 2026 don't rely on one smart model doing everything. They use a small team: one agent that plans, several that execute, and one that checks the work before it ships. That's agent orchestration, and you can run a DIY version of it in nothing but browser tabs.

Here's the trick: instead of dumping a big, messy task into one chat and hoping for the best, you split it into three separate passes, each with a clean, focused context (the sole info that chat window can see).

Planner pass: open a fresh chat, give it the full task, and ask it to break it into 3-5 specific subtasks another AI could execute independently.

Worker passes: open a new chat for each subtask. Paste in only that subtask plus the context it needs, nothing from the other subtasks. This keeps each pass sharp instead of confused by unrelated details.

Verifier pass: open one more chat, paste in the original goal plus every worker's output, and ask it to catch contradictions, gaps, or quality issues before you ship it.

Break this task into 3-5 specific, independent subtasks that a separate assistant could complete with no other context than what I give it: [YOUR TASK]

VERIFIER PROMPT:
Here's the original goal: [GOAL]
Here's what each subtask produced: [PASTE ALL OUTPUTS]
Check for contradictions, gaps, or quality issues before I ship this.

🎓 September 8

Build better 3D worlds with Astra

So GPT-6 Astra’s “Ghibli moment” might actually be Blender. Seriously, if you’ve seen one of the absurd Astra demos floating around lately, there’s a pretty good chance Blender is involved somewhere.

OpenAI has been showing Astra turning Blender models into playable worlds, and Matt Shumer has basically taken that idea and run directly toward “can I build GTA before GTA 6 comes out?” His increasingly ridiculous 3D experiments have now pulled 15M+ views.

So naturally, everyone keeps asking him the same thing: what prompt are you using?

Turns out, the crazy results are not coming from some magic prompt. Shumer says the real trick is the process around it.

His recipe is Reference → Assets → Assembly → Critique → Ship:

Start with real photos, maps, or generated concept art so Astra has an exact visual target.

Have specialized sub-agents build individual assets in Blender, then assemble them in Three.js or Unreal.

Give a fresh agent the render and reference image blind. Every difference it spots becomes the builder’s next assignment.

Keep looping until the critic can’t reliably tell which image is real.

For browser games, Shumer favors Three.js + Blender. For huge photoreal worlds, crowds, and characters, use Unreal.

Matt’s full walkthrough includes his starter prompt, /goal setup, optimization loop, and blind-testing method. Read his full guide here

At this rate, GTA 6 may indeed have competition from GPT-6 before launch day.

🎓 September 9

3 ways to use ChatGPT in a totally new way

The new GPT-6 Astra showed off some impressive capabilities, especially around voice and computer use. If you want to be like Corey over Labor Day weekend and vibe code on your phone while you’re out back barbecuing, follow these tips below!

First and foremost, if you haven’t tried the new ChatGPT desktop app yet, download it here.

Want to direct ChatGPT like the GPT-6 launch vid? Use Voice + computer use like so:

Open ChatGPT desktop, choose Workor Codex**in the top left, then go to →[Plugins](https://chatgpt.com/plugins?utmsource=www.theneurondaily.com&utmmedium=referral&utm_campaign=openai-s-1m-math-breakthrough-sparked-a-fight-with-anthropic)→ Computer Use. Install/enable it, turn on the Computer Use server + skill** toggles, and approve the permissions it asks for.

Start a task, turn on Voice, and tell ChatGPT to use Computer Use.

Then talk it through the job while it controls your computer: tell it to open apps, click, type, make changes, check progress, change direction, or explain what it’s doing. I do this like all the time.

Next up: want to direct it to do things, but on your phone? You need Remote control:

Start a Codex task on your computer, then enable Remote Controlin desktop settings and toggle “Allow Connections”to on.

Open the ChatGPT mobile app→ Remote and connect to that desktop session with a QR Code. Pro tip: you can do this with 2+ computers!

Now, your computer does the actual work while you steer it from your phone. Ever have a boss? This is how your boss feels making you do things.

Right click Open image in new tab to make this big enough to read

Uh oh! You just got vibe-code pilled and hit 0% remaining on your usage for the week! Been there too many times to count, friend. We got you

Do you have ChatGPT account from prior to GPT-6 launch? Use your Banked resets!

Go to Settings → Usage (sometimes shown as Usage & limits).

If you have a banked reset (or three), select the available reset, check when it expires, and click it twice to confirm using it.

A full reset refreshes your weekly Codex windows, then moves your weekly reset date seven days from now. Use sparingly… and follow Tibo (aka Mr Reset himself).

Much like drinking milk, please use in order of first to expire!

🎓 September 10

Give Meta Muse an ongoing job

OpenAI’s general agent tool (“Workspace Agents” currently) hasn’t been rolled out to regular people yet, so Muse might be the first real personal agent that many people use (we’ve heard Grokbot is great, but it costs hundreds of dollars to try).

If you’re interested in trying Muse, check out Meta’s 2-min Muse demo. Because Muse can keep working after you close the app, the best jobs for it are the ones you’d otherwise have to keep checking yourself.

Give it one ongoing job, like watching prices, planning a trip, or keeping you on top of an inbox.

Check Status to see what it did, what it’s doing next, and anything waiting for your approval.

If the job works, make it more proactive with Feed, Ideas, Goals, or the files Muse creates in Library.

The demo shows the whole loop: Chat, Status, Feed, Ideas, Goals, and Library.

What should you try first? Start small: give it one low-stakes job, then add permissions only when that job actually needs them. Just make sure you read those terms and conditions and make sure you’re cool with them first!

🎓 September 11

Turn your AI corrections into a personal benchmark

Public benchmarks can tell you how models perform on generic tests. Every’s new workflow turns the corrections you already make into a personal benchmark for your actual job.

Pick one recurring task and save the original prompt plus source files.

Review the first output and turn each correction into one yes-or-no check, like “one idea per slide.”

Grade the output yourself, then have AI grade the same checklist without seeing your scores. Tighten any check where you disagree, then rerun the same test across models.

Every found GPT-5.6 Luna beat larger models on many of Mike Taylor’s everyday tasks. That is the point: measure the cleanup you actually care about.

They’ve even got prompts for you to copy in the link above… definitely check it out!

🎓 September 13

Verify an agent’s memory before saving it

Persistent agent memory has a nasty failure mode: one bad conclusion can become “knowledge” that pollutes future work.

Microsoft researchers improved this issue by giving a separate memory curator read-only access to the environment before it saved anything. On CLBench, pass rate rose from 39% to 73% while task-agent cost fell from $3.38 to $1.68.

Let the agent propose a memory after the task.

Have a read-only checker verify it against the repo, docs, CRM, or other source of truth.

Save only the verified version, with its scope and source.

Before saving this memory, verify it using read-only access to [SOURCE OF TRUTH]. If confirmed, rewrite it with the exact scope and source. If it conflicts or cannot be verified, do not save it.

🎓 September 14

Microsoft Copilot Has a Secret Brain Swap

Buried inside Microsoft Copilot is a toggle almost nobody touches: which AI actually answers you.

Since earlier this year, Microsoft has quietly let Microsoft 365 Copilot users swap out the default OpenAI model for Claude, Anthropic's rival AI, on specific features. The clearest one is Researcher (Copilot's deep-research agent, which digs through your emails, files, chats, and the web to answer complex, multi-step questions). Different models write and reason differently, so if a Copilot answer feels flat, the fix might be a different brain, not a better prompt.

Open Researcher inside Copilot Chat and start a query.

Look for a model picker near the input box, sometimes labeled "Try Claude."

Switch it from the default model to Claude Opus, then rerun the same question.

Don't see the option? Ask IT to enable Anthropic models under Copilot settings in the Microsoft 365 admin center; it's off by default in the EU and UK.

Claude tends to write cleaner long-form analysis; the default model is often faster for quick lookups. Worth testing both before you assume Copilot just "isn't that smart."

Research [TOPIC], pulling from my emails, files, and the web. Compare [OPTION A] to [OPTION B] and give me a one-page recommendation with sources.

🎓 September 15

Create first, enforce the rules second

MIT’s new HardFlow method tackles a surprisingly common AI problem: when you force a model to obey every constraint while it’s still figuring out the answer, you can make the result worse.

Their technical method gives the model room to explore, then enforces the hard constraints on the final output.

You can borrow the same idea in your prompts:

Ask AI to solve or draft the best answer first.

Give it your non-negotiables: word count, required facts, formatting, tests, safety rules, etc.

Have it check and revise the final result against every constraint before returning it.