Parenting kids πŸ‘Ά and agents πŸ€– @tuikagents that reduce costs by up to 10x.

Visit my newsletter β†’
Joined September 2014
Multiple agents, multiple desktops. No more shared screen. Your Hermes Agent profiles, Grok Bots, and other agents all do their browser work in the same browser, share the same display, use the same cursor, and access the same filesystem. That makes parallel browser actions difficult because you end up with two agents fighting over the same mouse and window. @trycua Spaces fixes this by giving each of your agents its own desktop, so you can have several Spaces running in parallel. And a single agent can also have multiple Spaces. Why does that matter? > You can visualise what your agents are doing. > You can have two or more scrapers running at once. > You can get more done by running tasks in parallel. > You can protect each bot’s sessions from the others, since each has its own disk and browser profile. Everything still runs on the same machine, though, so don’t expect a huge productivity boost if your VPS only has 8 GB of RAM. Also, the free tier only allows two live sessions and 20 remote hours per month. So you won’t be running an army of agents performing browser interactions simultaneously without paying for it. In any case, this is a feature many users have been asking for: one Space per agent is now possible with any agent harness you can connect to MCP, including Grok Bot, Claude Code, Codex, Cursor, Hermes, and OpenClaw.
Your team and your Hermes can now share a computer. The new @trycua Spaces plugin gives Hermes its own Linux, macOS, or Windows Space on hardware you already own. Anyone on your team can open it, watch live, step in, and hand it back while Hermes keeps working. hermes-agent.nousresearch.co…
13
2
14
701
How many days does this take after paying for a subscription?
2
60
Just created my agency's 𝕏 account. I offer a custom AI agent to the first follower of @tuikagents πŸš€ Good deal, and it's not even Black Friday.
1
2
95
Chestnuts are easy to catch. But they are hard to peel. I learned something this morning while gathering chestnuts with my daughter. She was afraid of the spines, so I told her to pick only the ones that were already out of their shells. Suddenly, the job became fun and easy for her. This reminded me of coding. It used to be a delicate job. Good developers were often careful before shipping. Now, everyone wants to do it fast, like picking chestnuts that are already out of their shells. It's fast and easy: you take them home, and you cook them. But it turns out that the ones outside their shells may not be as fresh as the ones still inside. And after cooking, they're harder to peel. Exactly! AI-generated code is hard to "peel." It doesn't really matter if you're token-maxxing with Opus 5.5; there's a high chance your code will be messy and difficult to debug. Everyone says, "The first 90% of the website was easy to generate with vibe coding. The difficult part is the remaining 10%." That's partially true. Because if you have no idea how that first 90% was generated, you may have a lot more work on your hands. You just don't know it yet. I'm not saying we shouldn't use AI. Quite the opposite! But: 1. We must pay attention to every script that's generated. 2. Make sure we understand what we're creating. 3. Run automated tests on the codebase. 4. Run code refactoring skills. 5. Run code audits. If you do that, it will be much easier to "peel" the codebase. I love chestnuts. They're delicious, but peeling them can be a pain sometimes. Next time I go chestnut gathering with my daughter, we're going to do a little more work upfront for sure.
2
1
70
AI agents don't ride kick scooters. The so-called Superintelligence (SI) can't do what my 2-year-old daughter does. And that's because, no matter how intelligent an LLM is, it still needs a harness. Humans don't. > Humans fall. > Humans cry. > Humans say "no". > Humans get frustrated. And in the process, they learn. On the other hand, agents are not "free spirits": > They need a harness to handle memory. > They need a harness to connect to tools. > They need a harness to follow human instructions. > They need a harness to prevent them from taking malicious actions. That's why raising kids and building AI agents are two kinds of parenting. On one side, you need to provide freedom with a few guardrails. On the other, you need to be a "super-restrictive" dad. I'm both on the same day, and both are rewarding. For now, I'm just happy that I get to watch my toddler ride her scooter without respecting me at all, while my agents follow my instructions to the letter.
2
151
When you ask Grok Bot to access your LinkedIn to pull prospects and suddenly your account is blocked for 24h.
66
Gemini 4 Argon running on the Hermes Agent! Just do: /model gemini-4-argon --provider google
1
87
Gemini is here πŸŽ‰ after 2 weeks!
1
106