Job I handed Grok Bot this week: Clear the renewals and invoice reminders that keep bouncing back to me.
The agent got its own computer, a login I give it, a hard weekly allowance, and clear permission to pay only what was already on the approved list. Anything new or customer-facing waited for me.
I left for other calls. The Grok Bot agent worked the pile on its machine. I came back to approve the edge cases.
Hours leak when an email & social media task sit until you exit a meeting.
You can now hand those to Grok Bot (agents have their own computers, no slowing your PC down on a call, and you can use it from your phone!).
@bot is changing the game!
SMTP is the new prompt injection 🚨👀
Giving an agent email gives anyone a way to put instructions in its context.
That gets scary fast if your agent can also use files, send messages, access internal tools, etc.
A malicious email doesn’t really need to “hack” the system. It just needs to convince the agent to do something it already has permission to do.
The important takeaway from @adisingh: you can’t solve this inside the LLM
You need:
> permissions
> sender/recipient restrictions
> tenant isolation
> action limits need to be enforced outside the model
An agent inbox is basically an entry point into everything that agent can touch.
Stay safe.
THAT'S F**KING CRAZY, MY GROK BOT CONTENT DESK COSTS ME $20 A MONTH.
I GOT QUOTED OVER $15,000 FOR THE HUMAN VERSION.
I actually priced it out before I built it. Researcher, writer, designer, analyst, scheduler, editor – six roles, market rates, five figures a month before anyone ships a single post. Freelance instead of full-time? Cheaper, and then you own six calendars and six inboxes.
→ Seven agents replaced that line item. Here's the part that actually explains why.
One narrow job each. A model asked to be six people is mediocre at six things. Asked to be one, it's one. Six worse prompts beat one perfect one.
A fixed handoff order. The bottleneck was never ideas – it was the six hand-offs between having one and shipping it. Every hand-off is where a human team loses a day. Here it's a function call.
A seventh agent that only routes. Six agents without a chief of staff is a group chat. With one, it's a desk.
Nobody sleeps. Six people go home at six. The chain doesn't. Same week of work, compressed into an evening.
Zero code. Seven prompts. Ten minutes to stand up.
I still approve everything before it publishes – that gate I kept on purpose. And this hasn't been measured against months of results yet. When it is, the misses get written up too.
I wrote it all down so you don't have to guess at the prompts.
Seven agents, one article ↓
I am handing a Grok Bot agent this job this week: every time a tik tok posts, redistribute it to IG Reels, FB Reels, and YouTube Shorts.
It manages itself in my @NotionHQ second brain.
The charter is four short lines.
1. Source file from Notion on the agent computer
2. Caption to reuse
3. Upload to IG Reels, FB Reels, and YouTube Shorts.
4. Nothing publishes until I approve.
Run it once without hovering, then see whether the redistribute pile still waits on you.
Be careful with token usage though! My new usage bot might help: https://t.co/fhcPYMiD5B
Let me just say, I absolutely love Grok Bot @grok@bot!
Usage has been a real issue. Bots are not smart with their own usage and you can cause huge amounts of usage by accident. This is especially difficult when you need to get things done and have to start paying for additional usage.
The team should consider increasing usage and/or refunding a certain amount of extra usage for these first weeks. So much usage has been wasted by bots just doing too much and too often and users not knowing what all their bots are doing every second. I have had to spend quite a bit on additional usage to fully test my bots and much of the wasted usage was from bots doing more than I asked and more often than asked, duplicating routines, keeping unused browser tabs open and other things running on their PC, duplicate browser tabs, going above and beyond the call of duty, and generally doing too much.
Also, maybe there should be a dashboard that shows each bot's usage in real time and over certain time frames, as well as usage per task.
saved you all some tokens and made a usage bot: https://t.co/EUDU1NNfCo
send feedback if it works, i cant use it cuz I blew through enough tokens to get a slack message from the CFO asking me to set a limit
my number 1 tip for controlling your grok @bot usage is to avoid scheduled routines that run too frequently. for example, a 15 min routine runs almost 100 times a day, and every run consumes tokens. hourly or a few times a day is usually good enough.
the length of the chat with your bot can also make routines much more expensive. for recurring ones, try giving that to a fresh bot, while you continue your chat with your main bots (like a chief of staff).
give this post to your bot and ask them to help!
Job I handed a Grok Bot agent today. Title tags and metas for the site pages that still look empty in Search Console. It got its own computer with Chrome and the page list open, then kept going while I was on Zoom. Done means a draft sheet of new titles and metas. My gate is approve before any live edit.
Optional on-ramp if you hate juggling five windows. Firstmate is a chief of staff bot that can hand that same job to the agent for you (https://t.co/3w6RzjJ35O).