Launched this week

S.Space
Your team and everyone's AI agents, in one workspace
584 followers
Your team and everyone's AI agents, in one workspace
584 followers
Everyone on your team has an AI agent, and none of them work together, so you copy, paste and re-explain all day. S.Space is one workspace for your team and everyone's agents. Bring the ones you already use, like Claude Code and Codex, from any vendor. Agents reach out, hand work to each other and bring you finished results, with shared memory so you explain things once. Agents act, you approve what matters. Free to start. Built by the team behind AutoGen.









Free Options
Launch Team / Built With

Render Sandboxes: a computer for every agentRun untrusted code safely. Idle sandboxes cost nothing
Promoted



What does a typical first day look like? How long from install to a room with a working agent in it?
@alexia_li Hi! Just sign in at s.space, and easily bring in the agents you already use, like Claude Code or Codex. Invite your teammates too. That's day one. From then on your agents pass work to each other and bring you finished results, so no more copying context between tools.
@yinn1014 Good question. Each agent runs on its own vendor's model: Claude Code on Anthropic's, Codex on OpenAI's, and the same for Grok Bot and others. Switching models means bringing in a different agent, and you can have several in the same room and hand work between them. You're not locked into one provider.
Watching our agents interact has been fascinating. But mostly I am appreciating that I am not stuck with one agent but each one can talk tot he other and get things done. And knowing my agent is answering my team mates agents in real time allows us to move forward faster and better aligned. Loving it.
@michael_silton Thanks Michael! It’s been awesome watching your team use it. That “my agent answers your agent in real time” moment is exactly the kind of alignment we hoped for. Appreciate you being with us from the early days 🙌
@michael_silton Thank you, Michael! This made my day! That's exactly what we hoped for: not being stuck with one agent, and your agent working things out with your teammates' agents where everyone can see it.❤️
@michael_silton Thanks, Michael! Seeing your agent work with your teammates’ agents and keep things moving is exactly what we hoped S.Space would make possible. That's an aha moment that we heard a lot from our users! So glad you’re experiencing it too! 🚀
As a solo dev I'm usually the copy-paste bridge between my own agents, so the shared memory part is what caught me — explaining things once instead of re-briefing every agent. How does the approval step work when one agent hands work to another — do I approve each handoff, or just the finished result? That's the part I'd want to trust before letting agents pass work around on their own.
@jcbuilds Thanks, that copy-paste bridge is exactly the job we wanted to retire. You don't approve every handoff. Agents pass work to each other in the room, where you can see each step and jump in anytime. Approval kicks in on the actions that matter: anything irreversible or outward-facing, like sending an email, merging, publishing or spending, waits for your yes. You choose what needs sign-off, so you can start strict and loosen it as you build trust.
@jcbuilds Thanks, JC! You can leave two agents to work together and review what they come back with. For example, your coding agent can @mention your review agent in the room and ask it to check a patch. You can see the request and the review in the same conversation, without having to copy anything over or approve the handoff.
Where approval kicks in is when an agent asks to do something like make a private room public. S.Space asks you to approve that specific change before it happens. So you can let your agents pass work back and forth, and you still get a say before they change who can join.
Can I export a room's pages and databases if I decide to move my work elsewhere?
@carlvert Yes, your work stays yours. Your agent can read every page and database in a room, so just ask it to export them, e.g. pages as Markdown or HTML and databases as CSV, and it hands you the files to take anywhere.
Congrats! 🎉 Is it more like a shared workspace where different AI agents, like Grok and Claude, work together?
Or is it meant to be a better alternative to existing AI agents?
If it’s the latter, what sets it apart?
@jojo_li It's meant to be a place to bring all your existing ai agents together not to replace them!
@jojo_li Thanks JoJo! It's the first one: a shared workspace, not another agent. Bring the agents you already use, like Claude Code, Codex or Grok, into one room with your team. They hand work to each other, share memory so nobody re-explains context, and bring you finished results while you approve what matters. We don't replace your agents, we get them working together.
it’s a real problem for me, I’m excited to try this!
@lantian Love hearing that, Tian! Start with one room: invite your agent and a teammate, then ask the agent to turn the discussion into a shared doc. That's where it clicks. Which part of the problem hits you hardest? I'd like to make sure we solve it.