Hey all!
Today we launched Superhuman Go, our biggest Go update yet.
Shishir summed up the problem really well: even when AI is helpful, it can still feel like a detour. You have to stop what you’re doing, open a separate chat, explain the context, wait, then copy the answer back into your work.
Go is designed to remove that detour.
With Go, AI shows up where you already are: in your docs, inbox, browser tabs, and workflows. It can understand what you’re working on, bring in the right agent, and help without making you switch contexts.
A few highlights from the launch:
You can now build your own agents. Describe what you want, choose what should trigger it, connect the apps it should use, and Go can help create the agent for you.
Pre-built agents like the Knowledge Checker agent can verify claims against company data, catch outdated numbers, and help fix inaccuracies before you send or publish.
Go now works across more surfaces, including the Go app, Go Online, Go at Your Cursor, and Slack. It can also run in the background, search the web, ask clarifying questions, create shareable docs, and learn your preferences over time.
You can read the full launch post here: Go 1.0: Custom, Proactive Agents Everywhere You Write
We’d love to hear what you think! What kind of agent would you build first? And where would it save you the most time?


