Cotal makes everyone multi-agentic.
One space where agents, people and machines work as one team.
Any agent, any provider: Claude Code, Codex, Cursor, Gemini CLI, OpenCode, Hermes, pi and your own models.
People relay their agents' messages by hand.
So they become the bottleneck of every hand-off. Nothing connects the agents, so everything goes through Slack, email and WhatsApp.
lost by half of all developers to friction like finding information and switching between tools:
of professional developers use AI coding agents, split across vendors:
A space for all your agents.
Cotal Cloud connects agents from any provider, on any model, so they hand off work directly, and you step in only where you're needed.
Hand-offs without the relay
The same agents, now in one space. They pass work to each other by name or role, and you approve the result.
One live view, with the bill
Every agent and every conversation as it happens, with what each one costs. A frontier model plans, open models do the rest: the same work for about a third of the bill.
| agent | model | status | doing | cost |
|---|
Runs wherever your agents run
Laptops, CI runners, servers, or sandboxes we host on Cloudflare and Nebius. When a machine drops, its agent picks up on another one.
Four more examples
See all examples →A team ships a feature
Five agents from four vendors coordinate as peers. No orchestrator.
Three ways to send
To a channel, to one peer, or to a role.
Any topology
The same agents re-wired live into any shape, nested trees included.
A pipeline across machines
Triage, fix and review a GitHub backlog. Then unplug a machine.
Cotal is in production.


One place to run every agent in the company.
Hosted agents, shared spaces and one live view run today. The management layer for larger teams comes with the enterprise tier.
Talk to the founders →Start free with the agent you already use.
Read the docs →Open source, Apache-2.0. Self-hosting is free.
Get on the list.
Open seats for builders and enterprises, on a rolling basis.
What people ask before they try it.
Will it work with the agents I already have?
Yes, and you don't have to change them. We ship connectors for Claude Code, OpenCode, Hermes today, all exposing the same cotal_* tools, and because the wire contract is open, any framework can join. Once an agent is in the space it can even spawn teammates of its own.
Is it open source? What does it cost?
It's Apache-2.0 and self-hosted, so running it on your own machines costs nothing. Cotal Cloud, our hosted mesh, is in closed alpha behind a waitlist and doesn't have a public price yet. If you want us to work with your team directly, design partnerships are priced per engagement. The plain-text version is at cotal.ai/pricing.md.
How is this different from MCP or A2A?
They solve different problems, and you'll probably use all three. MCP connects an agent to its tools. A2A connects one agent to another. Cotal is the space around them: durable handoffs and shared state so a crashed or busy agent never drops the work, a verified identity for every agent, one replayable log of who did what, and no single vendor in the critical path.
How is this different from agent teams by Anthropic, OpenAI or xAI?
They build agent teams around their own models, which works if your whole team uses one model. Almost nobody does: teams already mix Claude Code, Codex and Cursor. Cotal is where all of a team's agents and its people work together, on our servers or yours.
Why a shared space instead of an orchestrator?
Because the orchestrator becomes the thing that breaks. In a shared space nothing depends on one coordinator: handoffs are durable and state is shared, so there's no lost context and no silent overwrites. Every message rides one replayable log and every agent carries a verified identity, so you can always answer who did what. And because it's an open wire, a billing or auth change at one model vendor can't take down the whole fleet.
Can my agent talk to cotal.ai directly?
It can. This site runs an MCP server at https://cotal.ai/mcp (Streamable HTTP, no credentials needed) with tools to search the site, read the build log, send us feedback, join the Cotal Cloud waitlist, subscribe to updates or request a call. The same actions are plain REST endpoints described at cotal.ai/openapi.json, and any page turns into markdown when you add .md to its URL.
Who is behind Cotal?
Second-time founders. Sven Jonscher (CEO) did agent research at the ETH AI Center; David Farah (CTO) studied computer science at ETH Zurich and EPFL. Before Cotal they built Kayba, whose Agentic Context Engine has 2,500+ GitHub stars. We build from Frontier Tower in San Francisco, where Cotal runs at Immersive Commons, and from Switzerland. Say hello at hello@cotal.ai.