DorkOS vs Grok Bot

Each chat picks its own agent tool. Start a job on Claude Code, run the next on Codex, keep a third on OpenCode, without leaving the tab or changing any setup. All three land in one list, and they keep running for as long as DorkOS does.

You write the job once and say when it runs: every night, every Monday, every hour. DorkOS starts the agent at that time on your own machine, then messages you when it finishes or needs a decision. Remembering is not your job.

Worth knowing before you switch one on: their own advice is to test a routine first, and their warning is that a test run performs real work. In their words it can navigate websites, change files and call connected tools, so a test is a real send rather than a rehearsal. The difference from ours is where the job runs: theirs in SpaceXAI’s cloud whether or not your laptop is open, ours on your own machine, which has to be awake.

An agent can propose a repeating job. It never starts one on its own. You get a card naming the agent, quoting its reason, and showing the next three run times and the exact instructions. One button runs it once, supervised, before you agree to it forever.

Your agents share rooms the way people share a group chat. They see each other, answer each other, and hand a job along. You set how often they may reply, per room, so a busy channel never turns into a runaway bill.

One thing to know about the shape of it, from their own security page. Every bot on your account uses the same cloud computer, so files and signed-in browser sessions are shared between them, and the page says plainly not to treat separate bots as a security boundary. It is many bots on one machine rather than many machines. DorkOS puts agents in shared rooms on your own machine, and caps how much they may say to each other.

Four dials cap the replies your agents may trade: how many in a row, how much of one conversation a single agent may take, how many per room each hour, and how many across every room each hour. Set them once, or room by room. No room can skip the dial that covers everything.

This is the deepest difference, and neither answer is wrong: they answer different questions. Grok Bot’s computer is the product. It is already set up, it holds your files and your signed-in browser sessions between jobs, and turning off what it may do on your laptop does not stop it working in the cloud. DorkOS has no cloud for your work to sit in. Your projects, your sessions and your history stay on your own computer, under the accounts already signed in there.

The phone app is a real one rather than a viewer: you can start work, answer a bot’s questions, approve steps and review results from it. They are straight about where it stops. Some advanced desktop controls and teach-by-demonstration are not on iPhone, and editing a routine’s schedule, changing a bot’s instructions, reviewing run history or deleting a routine all send you back to a desktop. It is iPhone only, not iPad or Android. So this row is close rather than level: both let you approve work from your pocket, and theirs asks you to finish some of it at a desk.

An agent that needs permission asks, and the question follows you: the browser on your phone, a Telegram or Slack message, or a notification banner on your Mac. Only the person you name gets asked, so it is never broadcast to every chat you have connected.

One bell holds every question, approval and finished run, with the things waiting on you above the things that merely happened. What you have read stays read on your other devices. A question left unanswered for the few minutes you chose rings your phone.

DorkOS does not replace a marketplace of editor add-ons built up over years. It adds the other half: package a working agent, with its instructions and its tools, and hand the whole thing to someone else in one command. Signing those tools in to outside services is the part still finding its feet.

Skills are reusable instructions a bot keeps and reuses, the same idea DorkOS builds on, and connectors are installed for the whole account. Its overview page adds that a bot can use connectors and MCP tools where available, so the common standard for reaching outside tools is in the picture. The half we could not find is sharing: nothing in its documentation describes publishing a skill for other people, or installing one somebody else wrote. That may simply be early, which is why this row is "partly" rather than a yes.