Make a supply plan that balances coverage, cost, and stock.
Change the briefRun working toolsInspect the result
YOUR TURN TO EXPERIMENT
Free to explore · No sign-in or API key
Give the agent a brief.
Prepare a fundraiser supply proposal from a fictional catalog. Maximize complete drink-and-snack packs, compare spending preferences, and verify the total including delivery.
Cover a group of 30 within a $65 budget.
Ready when you are
The tools calculate and check your inputs in an isolated JavaScript runner.
Working local tools. Authored controller and classroom data. No live AI model or external action.
A SMALL BRIEF. A VISIBLE RESULT.
See the agent at work.
Start with a scenario, change a detail, and run the demo. Your result and the steps behind it will appear here.
1 · Read the brief2 · Use tools3 · Check the result
Try this twistFirst, compare spending less with including more fruit. Then find the smallest whole-dollar budget that covers 30 people with a $5 delivery fee.
Open the code behind this agentRead it, change it, run your own variation.
This is the JavaScript that powers the demo. agent.js coordinates the workflow; tools.js contains its available tools. Tool calls and returned data are inspectable after each run.
Ready when you are
⌘ / Ctrl + Enter runs your agent.js draft with the current inputs. Tab leaves the editor. Package imports and network calls are unavailable in this bounded environment.
The real research behind this demoWhere the idea comes from, and what this classroom version does.
Anthropic and Andon Labs' Project Vend examined inventory and pricing decisions in an office shop. Its reported failures motivate the explicit cost, stock, and budget checks you can inspect here.
The basket search and arithmetic execute locally over invented prices. It places no orders and makes no investment or profitability claims.
STEMtiq authored this exercise. The named organizations do not endorse it. The JavaScript tools run locally; the controller is programmed, and classroom records are labeled examples. This is not a live instance or reproduction of the referenced research system.
Anthropic · Andon Labs
Project Vend: Can Claude run a small shop?
A real office-shop experiment gave Claude inventory, pricing, search, and communication capabilities, exposing failures such as pricing below cost.
The first reported shop did not make money. Our bounded basket search teaches explicit cost and stock checks; it does not reproduce the experiment or promise business success.
Your JavaScript really runs. The model decisions and school data are authored simulations, so you can learn without an API key. Every workspace also includes a separate real SDK example to explore next. Passing the lab’s cases is practice, not proof that an agent is ready for real-world use.