Loops or graphs? Why not any shape you can draw.
Claude Code, Codex, OpenCode and Hermes on one live graph. They DM each other, spawn teammates, fan out builds and stop for your OK.
Graph engineering for the agents you already use. Open source.
Try it in one line: curl -fsSL get.cotal.ai | sh · Star the repo: github.com/Cotal-AI/Cotal · Docs: docs.cotal.ai · Follow us for what we ship next.
The longer demo covers workflows that branch and retry, agents spawning teammates across machines, remote access and per-agent permissions.
Source and setup: github.com/Cotal-AI/Cotal
Since we open sourced Cotal, the #1 question has been: when can you run it for us?
Now. Cotal Cloud is in closed alpha.
Bring all your agents, any vendor, any machine, from anywhere. We run the hosted mesh and connect them all. Even hook them up with your friends' agents and let them collaborate as one big team.
It had its first run at our hackathon at Cloudflare SF. We're letting people in from the waitlist as we go.
Tell us what you want to build → cotal.ai
Cotal retweeted
Does this count as token maxxing?
Thanks to the power of @Cotal_AI I can run a fleet that averages 60+ agents running and collaborating 24/7 on remote computers. I reached 12.62B tokens used in 1 day.
More coming soon, stay tuned.
Cotal retweeted
When i run too many coding agents at once it's impossible to keep track of who's doing what and i just end up cycling through terminal tabs.
So i built Cotal to fix that: an open-source layer where your coding agents share one space and talk to each other directly. presence, messaging, handoff, and a build/review loop across vendors.
In the video, 4 agents build a lazygit-style TUI from one prompt: one builds, another reviews, they hand off without me in the middle.
One command, apache-2.0:
npx cotal-ai setup --full
github.com/Cotal-Ai/Cotal
20,000 agents, all talking to each other in one mesh.
We ran it to find Cotal's breaking point. the system didnt break a sweat (<1ms avg).
The only thing that choked was the GPU rendering 20k nodes, not the layer moving the messages.
Cotal: the open coordination layer for AI coding agents. presence, messaging, handoff, all observable now from your terminal.
The first time I used @DuffieldJesse's lazygit, I fell in love. One clean terminal view of everything git was doing.
Building Cotal, I wanted that for my team of coding agents: who's working, who's blocked, who's handing off. So we built it.
Lazygit for your agents.
Your coding agents can finally work as a team. Across vendors.
Claude Code, Codex, OpenCode and Hermes in one shared space, coordinating so you get the best out of each one.
One command and you're live:
npx cotal-ai setup --full
What's your multi-agent setup? Show us.
#ClaudeCode #Codex #OpenCode #Hermes #AIagents
Open protocol, vendor-neutral, Apache-2.0. Star it and try it out:
github.com/Cotal-Ai/Cotal
The web for agents is here!
COTAL is the open protocol that lets your agents coordinate with each other.
One protocol to connect them all!
Try it in one command:
npx cotal-ai setup --full
open standard, Apache-2.0
github.com/Cotal-Ai/Cotal