Grok Bot 1.2 can send texts now. I tried it 📱
It sent an iMessage from my Mac, as me, after an approval card.
Great for texting people you already know.
But think about whose number the other person gets 👇
The plumber gets your personal cell.
And your Mac has to be on for your Bot to send.
The new number you can text your Bot at? It only answers your own phones.
That's why my Phone Bot still runs on @usebland.
Bland gives your Bot its own number.
Anyone can call it or text it, plain SMS included.
Bland is the phone line. Your Grok Bot still does the thinking.
It works with US and Canada numbers.
Native texting is your number. Bland is your Bot's number.
Test calls FREE with Bland credits 🆓 Its own number with calls and texts is $9.99 the first month, then $19.99/mo.
Save this before you give your cell to a contractor 🔖
FREE templates below 👇
Buried 58 minutes into Friday's Grok Bot livestream: the line that gets your X account banned 🚨
Someone asked if letting Grok Bot send DMs gets you suspended.
@mattyp: No. The built-in X integration asks for your OK before it sends. He says that's in line with the terms.
Your own scripts or sketchy stuff? Banned 👀
Would you let your Grok Bot touch your X account?
🚨 FREE USAGE JUST DROPPED 🚨
Your Grok Bot week should last ~7% longer now. You didn't touch a thing.
Routines got the biggest bump: up to ~15% more 🔥
7% of a week is about half a day of extra runway. (My math, not theirs.)
Run a lot of routines? You win the most here.
Don't take anyone's word for it:
📸 Screenshot your usage % right now
📅 Check it at the same point next week
👀 Compare the two
Then you'll KNOW if it hit your account.
Not the full fix. But they keep shipping 👇
Your Grok Bot usage might be melting in Cursor 🫠
Cursor and Grok Bot pull from the same weekly pool. And on "Cursor Default," Cursor can move your new chats to the newest Grok, even after you picked a different model.
Lock your pick:
Cursor Settings
Agents → Conversation
Default Model → Last Used
Restart Cursor after.
Then check the dashboard switch below so your Bot's coding jobs aren't running on Fast 👇
Save this before your next reset 🔖
Your Grok Bot can now read your texts, search your Mail and write in your Notes. On your Mac 🤯
Messages, Mail, Notes, Reminders, Calendar, Contacts and Location. Each one gets its own Allow button.
Now stack it with Tailscale, which landed in the same desktop app this week 🔥
Put a Mac at home on your tailnet, run your own MCP server on it, and your Bot can reach it without opening it up to the internet.
Your apps on your desk. Your own tools at home. That's the combo I'd set up this weekend.
Real work is rarely one folder. API repo, web repo, infra repo, one product.
hermes project create "My App" ~/code/api ~/code/web
Desktop groups sessions by that name. bind-board gives every kanban task a worktree and a branch convention.
Approving the same command for the third time this week?
hermes approvals suggest
It mines your past yeses and proposes an allowlist. You pick. Nothing writes itself.
If a model id 404s, check the id before you check your keys.
Providers know models as vendor/model. Drop the prefix and it looks like an outage.
hermes model picks the prefixed id for you. Keep the habit anyway.
A long Telegram thread pays for the whole backlog every time you come back.
hermes config set compression.idle_compact_after_seconds 1800
Size-based compaction waits for the window to fill. This one fires on 30 minutes of silence.
🚨 Grok Bot users on $200 Ultra are getting squeezed twice this week.
First the weekly pool. One user went to bed at 28% and woke up near 60%.
Now users are reporting their Bots refuse local Claude Code and Codex, push them to Cursor Cloud, and some hit a demand for credits.
So you burn the pool. Then you pay again to code.
Who thought this was a good move?
Claude Code and Codex win either way. People will just run them without the Bot.
Let our Bots use the tools we already pay for. And show us what's burning the pool.
It's time Grok Bot users get heard.
Switch models mid-conversation and the next turn rereads the whole thread at full input price.
The cache key includes the model. New model, cold cache.
Batch the switches. Or start a fresh session. Subagents get their own context. Ping-ponging a long thread is how the bill shows up.
Hermes can save a fact to MEMORY.md mid-session and still not use it in the next reply.
The write hits disk immediately. The copy in the system prompt loaded at session start, on purpose, to keep the prefix cache warm.
Need it now? New session, or hermes -c on the next turn.
Every web page Hermes reads can burn a full-price call to your main chat model.
web_extract's summarizer defaults to that model.
Route auxiliary.web_extract to a cheap model, or use the browser tools when you need the raw page, not a rewrite.
A monitoring job that talks every run trains you to ignore it.
If the cron response contains [SILENT], Hermes delivers nothing.
"If nginx is healthy, respond with only [SILENT]. Otherwise report the issue."
Uptime, disk, backups. Speak only when something is wrong.