Argon — Git for MongoDB

Branch, time-travel, merge and undo your MongoDB. Any driver, real mongod,
versioned history underneath. Built for AI agents.

Three ideas, thirty seconds:

1. A branch is a pointer, not a copy — created in milliseconds at any
   data size.
2. checkout turns a branch into a real MongoDB database — pymongo,
   mongoose, mongosh, indexes, aggregation, transactions: all real, and
   every write becomes versioned history.
3. Nothing is ever lost — diff it, merge it, undo it, rewind it, or pin
   it forever.

Install

brew install argon-lab/tap/argonctl      # macOS
npm install -g argonctl                  # cross-platform

# MongoDB must run as a replica set (one-node is fine):
docker run -d --name argon-mongo -p 27017:27017 mongo:7 --replSet rs0
docker exec argon-mongo mongosh --quiet --eval 'rs.initiate()'

The flow

main ──branch──▶ experiment ──checkout──▶ mongodb://…  ← any driver
                                              │
                     ┌── argon diff ──────────┤  every write captured
                     ▼                        ▼
        merge (a data PR)          or   undo / discard / rewind

# 0 · Bring your data in ("git clone") — or: argon projects create myapp
argon import database --uri mongodb://localhost:27017 --database myapp --project myapp

# 1 · Branch — instant, no copy
argon branches create experiment -p myapp

# 2 · Get a real database for it, capture writes
argon checkout -p myapp -b experiment      # prints a connection string
argon watch    -p myapp -b experiment      # keep running while you write

# 3 · Review and merge back — a data pull request
argon diff          -p myapp -b experiment
argon merge preview -p myapp -b experiment
argon merge apply <plan-id>

# …or rewind instead of merging
argon restore reset -p myapp -b main --time 2026-07-07T09:00:00Z --backup pre-incident

Prefer clicking? argon console serves a local web console (UI + REST API)
and opens your browser.

What you get

Storage stays bounded: snapshots + retention-window GC keep state plus a
window of history, not every write forever. Details and the consistency
model, stated honestly: ARCHITECTURE.md.

For AI agents

claude mcp add argon -- argon mcp        # 13 tools: sandbox, diff, merge, undo, pins
pip install argon-agents                 # LangGraph checkpointer + Mem0, over REST

saver = ArgonCheckpointSaver.from_sandbox(argon, "myapp")   # checkpoint on a disposable branch
saver.merge()                                               # adopt the run — or .discard()

argon.create_pin("myapp", "eval-v1")                        # pin the dataset once
run = argon.sandbox_from_pin("myapp", "eval-v1")            # identical state, every eval run

The full agent workflow: docs/AGENTS.md.

Documentation

Community

Issues ·
Discussions ·
Contributing ·
argonlabs.tech