How it works
Make your first data review
Open the live console with sample orders and your own temporary session. Run the two-agent scenario: the script accepts planner’s proposal, then you inspect executor’s conflict and undo the rejected change. Continue locally to drive every step yourself.
The anonymous console exposes sample-data review tools. Native database connections require a local deployment. Your demo session expires automatically.
One order. Two agents. One decision.
Review changes to your business data
Conversation checkpoints do not capture changes to external orders and prices. This example makes the MongoDB data itself reviewable.
01 · Pin the input
Freeze orders at baseline: order-1 costs $49. Both agents fork that same pin, so their starting data is identical.
02 · Compare proposals
Planner changes the price to $44. Executor proposes $1 on its own branch. Both use ordinary MongoDB driver writes.
03 · Review before adopting
Inspect the planner diff and apply its merge plan. Main now contains $44. The executor plan surfaces a real conflict against that accepted change.
04 · Reject and recover
Undo the executor proposal on its branch, then discard it. You can also undo the accepted merge to restore $49; that recovery is recorded in history.
Run the complete Python example with the local quickstart. It asserts identical input, a real conflict, accepted data and the restored result. No LLM key is required.
Explore the concepts
This illustrated walkthrough is a simulation; the console above runs the real engine.
The full loop, from branch to data PR — the terminal drives, the panels follow. · simulated
step 1/9· Create a project — capture records supported document writesauto-playing
argon · cli
simulated
$ ▍
mongodb://localhost:27017 · shop
database view
// your existing database, before Argon
products (3 documents)
{ _id: "s1", name: "trail shoe", price: 49 }
{ _id: "s2", name: "road shoe", price: 89 }
{ _id: "s3", name: "track spike", price: 74 }
…
history graph