Open source · MIT · self-hosted
Git for MongoDB,
built for AI agents
Give every AI agent its own branch of your database. Review what it wrote, merge what works — discard or undo the rest. Nothing touches production.
$ brew install argon-lab/tap/argonctlthe whole idea, in one loop
$ argon sandbox create -p prod --ttl 1h
the agent gets its own branch — a real MongoDB database, prod untouched
01 — Inert by design
Named after the noble gas for a reason
Argon — element 18 — is inert: it doesn't react with anything. That's the contract: experiments happen on branches, history is append-only, production is never touched.
0.86 ms
to create a branch — p50, measured on a 50k-entry project
479 B
of storage per branch — branches are metadata, not copies
diff → merge
adopt an agent’s work as a reviewed data PR — or undo it, per-actor
numbers from the public benchmark suite — reproduce them with docker compose up
02 — Familiar workflow
The Git muscle memory you already have
| git (code) | argon (data) | what it does |
|---|---|---|
| git init | argon projects create | start versioning |
| git branch feature-x | argon branches create feature-x | isolated branch in milliseconds |
| git checkout feature-x | argon checkout -b feature-x | a real MongoDB database + URI |
| git log | argon time-travel info | see every past state |
| git diff main | argon diff | document-level three-way diff |
| git merge + PR review | argon merge preview/apply | reviewable data PRs, exactly-once |
| git reset --hard | argon restore reset | rewind mistakes |
| git revert | argon undo --actor agent:x | revert one agent’s session |
agents can drive all of this themselves via MCP: claude mcp add argon -- argon mcp — see the agent workflow
03 — How it works
One idea, carried all the way through
your app · your agents
any driver, unchanged
every write →
append-only log
pre/post images · actor · LSN
views →
branches
— real MongoDB databases
time travel
— any state, by LSN or time
undo
— per-actor, conflict-safe
04 — Built in the open
Claims you can check
Deterministic replay, property-tested in CI — same history, same state, byte for byte
tests →Every number on this site reproduces with docker compose up
benchmarks →MIT licensed, self-hosted — your MongoDB, your data, your infrastructure
license →Postgres has Neon. MySQL has PlanetScale. SQL has Dolt. Data lakes have lakeFS. MongoDB has Argon.
Branch your MongoDB today
Open source, MIT licensed, self-hosted. Star the repo to follow along — or run the benchmarks yourself first.