A few weeks ago I covered what Grok Bot is. Since then people have pointed it at their inboxes and at the stock market, and the results have been fun.

This edition is the good stuff, the best builds, and a simple way to try it yourself.

But first, some catchup:

A free anonymous model, 0x Alpha, is topping coding tests.

Nvidia has bought a stake in power firm Cloverleaf.

ChatGPT on Mac can now read and send your texts.

Grok Lite briefly started replying in gibberish.

ShogunAI: a local Mac assistant that remembers your work and acts on it inside your apps

Wizstar: turns product photos and links into avatar videos and ads without filming

Domo: a family calendar you update by texting its own phone number

ScrollToll: counts every short video you watch and blocks the feed once you hit your limit

CoachAI: uses your phone camera to count reps and flag bad form while you work out

Grok Bot went public and people immediately wired it into everything from plumbing dispatch to the stock market. Here are the ones I found cool.

Jon O'Neill runs a plumbing company with no engineers.

In a single day he wired Grok Bot into his work-order intake and job booking across 6 different tools, from Gmail and Slack to ServiceTitan.

Lenny Rachitsky, built a small team of bots to cancel forgotten subscriptions off his card statements and answer his support email.

Mike P set a bot loose on 90,000 emails across 2 Gmail accounts with 1 instruction, clean it up.

It worked through the night and archived about 14,700 threads without deleting a single one, so anything important stayed searchable.

Liam gave his bot its own email address. Now it can verify accounts and send and receive its own mail, which removes the last thing that still needed a human in the loop.

Danny Limanseta, a game dev, pointed a bot at his own art tool and asked it to fill in the placeholder graphics in his game.

It read the project, wrote its own prompts, and produced 74 finished, game-ready assets in about 2 hours.

Farzad built a bot that runs a live stock-trading experiment and maintains its own public website, refreshing every 15 minutes.

It shows a portfolio up about 6 percent and ahead of the S&P by roughly 3 points. Read the small print though.

It started with $500, so the whole profit is about $30.

Curious enough to try it? Here is the short version.

Step 1. Install it. Go to x.ai/bot and download the app for your computer. It runs on Mac and Windows, and there is an iPhone app. No Android yet.

Step 2. Sign in. Open the app and click Get started. It signs you in with a Cursor account, not your X account. Finish the login in the browser window that opens, then come back to the app.

Step 3. Create your first bot. Give it a name, one clear job, and a short description of how it should work. Keep the job narrow.

Step 4. Give it a task, then log it in. Type what you want in plain English. A good first job is one that touches something you already use, like your inbox. Try this:

When the bot reaches the Gmail login, it hands control back to you. You type the password and the 2FA code yourself, then hand control back.

Keep approvals on for anything that sends, pays, or deletes, and start with one low-stakes job before you trust it with more.

I also recorded a full walkthrough building 10 bots from scratch, if you would rather watch first:

Not much today tbh. People are having fun again, and it helps that the people shipping this are on a tear.

Grok Bot is a joint effort from xAI and Cursor, launching alongside a fresh Grok model. Say what you like about them, they are not standing still.

So now go and give your bot a silly name and a boring task you have been putting off, then see what it does. Reply and share your best ones with me.

Until next time,
Vaibhav 🤝🏻

If you read till here, you might find this interesting

#AD 1

More B2B companies are moving to usage-based and hybrid pricing — and finance teams are feeling it. Revenue recognition gets messier, forecasting gets harder, and the manual work compounds fast.

Tabs and PwC teamed up to break it down. In this on-demand session, Rebecca Schwartz and Amit Dhir share how leading finance teams are handling the operational reality of dynamic pricing models — and where AI fits in.

Watch the recording for concrete examples, a practical rev rec framework, and a clear-eyed look at what it takes to scale without the overhead.

If your team is navigating this shift, this session is worth your time.

Watch the Recording

#AD 2

Watch Fin and Plaid now on demand to see how AI can resolve issues directly in customer conversations. This includes everything from reducing bank-linking friction, enabling proactive support, and helping customers complete real financial tasks.

Watch now