Hey everyone, another question from the Velozity team as we get closer to launch.
Last time we asked how AI work moves between people on a team. This time we want to dig into trust.
Right now most people keep their AI on a short leash: it drafts, it suggests, but a person clicks send. That's safe, but it also means every handoff between a person and their AI, or between two people's AIs, still needs a human in the loop to move things along.
We're building Velozity so agents can work more directly inside a team's shared space, chat, tasks, files, calendar, with a person approving anything that leaves the workspace rather than approving every single step along the way.
Is there a way to see everything an agent did while you were away?
@jasmine_charles Yes. Everything an agent does is posted in the space it worked in, so the full record is there when you come back: what it was asked, what it did, and the result.
If you don't want to scroll, you can just ask "what did I miss?" and get a summary of what happened while you were away, across your spaces. There's also a morning brief that covers the same ground each day.
The shared AI agents idea is cool… how does the approval step work for things like payments or contracts?
@tery_emilson Thanks, Tery. Short version: agents prepare, a person approves, and money and signatures stay in human hands.
Nothing goes out without an approval card. When an agent wants to send an email, post, or submit a form, it stages the exact text and the exact destination on a card. Nothing happens until that is approved.
Only the agent's owner can approve. Teammates in the shared space can see the card, but their click is refused.
An approval covers one action only. It is tied to those exact words, that recipient, and that page. If anything differs at the moment of sending, it blocks and asks you again.
Payments and contracts are deliberately off limits for agents acting alone. Even a routine you have allowed to run automatically is stopped at any button like Pay, Buy, Checkout, Sign contract or Accept terms. Billing, cards, and subscriptions always need a fresh card, however clearly the agent was told.
Payment apps are read-only. With Stripe connected, an agent can look up invoices and revenue but cannot charge, refund, or pay out.
So today an agent can draft the contract, chase the invoice and set up the follow-up, and you do the paying and signing yourself.
@stormshot Yes, we built on open standards wherever one exists, so nothing here is a proprietary lock-in:
Tools for agents: Model Context Protocol (MCP). Everything an agent can do inside Velozity (read a space, post a message, drive the browser) is offered to it through MCP, the same standard Claude Code and Codex already use.
Talking to agents: Agent Client Protocol (ACP). Gemini, Kimi, OpenCode, and Cursor agents run over ACP. Claude Code and Codex are run through their own official command-line tools, unmodified.
Sign-in and connected apps: OAuth. You log in on the provider's own page, and we never see your password.
Calls and screen sharing: WebRTC.
If you meant a specific standard, such as a security certification or a data regulation, tell me which one and I'll give you a straight answer.
The difficult part is transferring work from one person/agent to another. Is it possible for me to delegate a task to an agent through a chat conversation or through a call transcript and get responses on the same thread?
Btw, Congratulations @ishudarshan and @see_akash 🎊✌️
How does the autopilot feature decide when a page variation has enough traffic data to pick a clear winner?
@sansa_grey Hi Sansa, thanks for stopping by! Velozity doesn't do page variations or traffic testing. It's a shared workspace where your team and AI agents work together. Happy to answer any questions.
How much time does your team spend repeating the same context to different AI tools?
Congrats @ishudarshan & team!
Thanks for sharing @rohanrecommends!
@hamza_afzal_butt Thanks, Hamza. Honestly, it varies, and sometimes it still takes longer than you'd expect. What's changed for us is that there's no back-and-forth anymore.
Everyone works in the same space, so it feels like one ongoing conversation, a bit like a WhatsApp group where the whole team and their agents are talking together. The agents feel like teammates, and everybody stays in the loop the whole time.
The part I like is that agents sit alongside my teammates, so I don't have to move work between tools by hand.
@shreya_verma12 Thank you, Shreya! That's the part we care about most. Moving work between tools by hand is a hidden tax on every team, and it disappears once agents sit in the same place as your teammates. Which handoff has saved you the most time so far?