Shared memory.
One memory for the pair, encrypted on your machine, distilled after every run so it gets sharper instead of longer.
Sno Station
Agents, assemble.
Two heads are better than one, and they get better with every job.
Never stops at the cap. Never grades its own work. Learns from every job.
- Apache-2.0
- assembled in public
- runs on your laptop, no daemon
- harnesses Claude Code · Codex · OpenClaw
Product
Sno Station
Agents, assemble.
Two heads are better than one, and they get better with every job.
Never stops at the cap. Never grades its own work. Learns from every job.
You already pay for Claude Code and Codex. You already run them side by side, two terminals, and you carry the results across by hand. Sno Station is the place where they stop being two tools and become one squad.
It runs on your laptop. No daemon, no server, no cloud required. One shared memory they both read and write. A mailbox they use to wake each other up. A set of skills they follow, and rewrite when they learn something. Apache-2.0, edge to edge.
Four things inside:
Agent communication.
Sno Reach: either agent can call the other, hand off a task, or ask for a review. Plain files, no daemon.
Skills for the squad.
Peer review, handoff, orchestration, reflection, assembly. Written to be followed by any agent in whatever language you use with it.
The RSI loop.
Every night it reads the day's sessions, proposes changes to the agents' own skills, and waits for you to say yes.
Why Sno
Agents that get better with every job they finish.
Claude Code and Codex stay the same. The workplace they share does not. After each job it rewrites its own skills, trims its own memory, and has a stranger check the work. What it learns is yours to keep. Your masters, your rules.
Every job makes the next one better
1. The skills rewrite themselves. You sign off.
After a job the agent edits the skill it tripped on and sends it as a pull request.
2. Memory that gets sharper, not longer.
After each run the shared memory is distilled. Fewer lines, each one earned.
3. Nobody grades their own work here.
A third agent from another lab checks it. It reads the diff, not the name on it.
4. A fix counts when the next run proves it.
Yesterday's change is measured against today's. The number decides if it stays.
Two heads are better than one
5. Not a second opinion. A second lab.
Agents from one lab share blind spots. A review from the other lab has never come back empty.
6. One hits its cap. Rotation puts the other in.
Rotation: the other agent takes the same line, same memory, same branch. The job doesn't stop.
7. Catch the mistake before it gets expensive.
A wrong plan caught at step three costs three steps, not forty. Two agents cost less than one that has to redo the work.
8. Different minds, different jobs. That's Dispatch.
Dispatch: one explores, one finishes, and each gets the work it's good at.
How it works
Connect once. Then keep working the way you already do.
Sno Station sits under the agents you already run. You don't switch tools. Three things change.
1. Work as before. The squad is there when you need it.
- One of them hits its cap. Say
sno reach callfrom the other one. It picks up the task from the shared memory and the mailbox, context intact. That's Rotation. - You want a second pair of eyes. Ask either agent to
peer-reviewthe other's work. The reviewer is always from the other harness. - You have a job with two halves. Let one plan and one build, or one build and one review. That's Dispatch.
2. Every night, it learns
The RSI loop reads the day's sessions, finds the mistakes that repeat, and proposes changes to the agents' own skill files. In the morning, rem-reflect accept <id> for the ones you like. Nothing changes without that accept.
3. Get started
Say Sno onboarding inside any Claude Code, Codex or OpenClaw conversation. The agent runs the setup itself through the sno CLI: shared memory, Sno Reach, the Duo skills, and the hooks each harness needs. No package names to remember.
(Until the onboarding skill lands: "Not yet. Watch the repository." Do not publish this step as live before the README does.)
What you get:
- A memory that survives the session and gets sharper with use.
- A review that comes from a different lab, so the blind spots don't line up.
- A squad that keeps working when one member runs out of quota.
- Skills that carry what the squad learned, in plain files you can read and edit.
Squad
Solo. Duo. Squad.
A squad is any number of agents with roles, sharing one workplace. A Duo is the smallest one, and it's where most people start.
Four things in the box, above. Four things the pair can do, below.
Two agents from two labs, one shared workplace. That's a Duo. It does four things a single agent can't:
They complement.
One plans wide, one builds tight. Each takes the part it's good at.
They check each other.
The one that didn't write it reads it. Both directions.
They teach each other.
One reviews the other's run and distills what to change. That's where the nightly loop gets its material.
They can take orders.
One agent leads, others work under it, and the lead decides. For a long plan that won't fit in one context, that's the shape that holds.
Add a third seat and you have a grader who never wrote the work. Add more and you have a squad. The rules are the same at every size: one workplace, plain files, no daemon, and a human who signs off on anything that changes the rules.
We never say which agent is the careful one and which is the fast one. It flips by month and by job. The point is that they differ.
Security
Yours, and it stays yours.
Everything the squad learns lives in one workspace on your laptop.
- No daemon, no server, no cloud required.
The product is complete without a cloud side. When a cloud side comes, it's optional.
- Encrypted from first use.
The memory store is encrypted on your machine with a key that is provisioned once and never leaves it. Sno never receives your database or your key.
- Plain files you can open.
Memory, messages and skills are files in a folder. Read them, edit them, export them, delete them.
- Nothing changes rules without you.
The nightly loop proposes; a human accepts. The agents' own skill changes arrive as pull requests.
- Apache-2.0, edge to edge.
Everything that runs on your machine is open source.
- The full boundary, what it protects against and what it does not, is in
docs/security.md. Report issues to security@sno.ai.
FAQ
What is Sno Station?
Open-source software that turns the AI agents you already run into one squad on your own machine. One shared memory, agents that can call each other, a skill set they follow and rewrite, and a nightly loop that makes the whole thing better with use.
Which agents does it work with?
Claude Code, Codex and OpenClaw today. Any agent that can run a shell and read a skill file can join a squad; those three are the ones we run and test.
How is this different from the official Claude-to-Codex plugin?
That plugin is one-way: one agent is the boss, the other is a tool it calls. In a Duo both agents are equal. Either can wake the other, either can review the other, and the work carries across in a shared memory, not a copy-paste.
Do I need a server or a daemon?
No. Sno Station is plain files in a workspace on your laptop. There is nothing to keep running.
Where does my data live?
On your machine, encrypted, with a key that never leaves it. Nothing is sent to Sno. Export, edit or delete it whenever you like.
Will it change my agents' behavior on its own?
No. The nightly loop proposes changes to the agents' own skills. You accept or reject each one. Nothing changes without that accept.
Is it open source?
Yes. Apache-2.0 for everything that runs on your machine.
What does it cost?
Sno Station is free. A hosted version, for a squad that has to keep running when your laptop is closed, is coming; the product on your machine stays complete without it.
What is a squad? What is a Duo?
A squad is any number of agents with roles, sharing one workplace. A Duo is the smallest squad: two agents from two labs. Most people start there.
Which one is the smart one?
We don't say. It flips by month and by job. The point is that they differ, and that the one that didn't write it is the one that reads it.
How do I start?
Say Sno onboarding inside your agent once the onboarding skill ships. Until then, star the repository and watch it; it is being opened one piece at a time.