Skip to content
agentusehub

Order groceries online with a local LLM skill

The author built a grocery skill for a local Qwen 3.6 27B model that knows their fridge contents, past orders and real-time stock, and it ordered products for them online.

Time saved
Not reported
Setup time
Not reported
Running cost
Not reported
Cost range
Not reported

Use case overview

The author built a grocery skill for a local Qwen 3.6 27B model that knows their fridge contents, past orders and real-time stock, and it ordered products for them online.

Reported experience

  1. What they did

    The author built a grocery shopping skill for a local Qwen 3.6 27B model. It gives the model a loose picture of what's in their fridge, about 10 representative past orders covering preferences and local shops and SKUs, and real-time in-stock information. With that in context, the model placed an online grocery order.

  2. How it went

    The order went through while the author was cooking pasta, and they say the local model fully solved the real-time stock problem that cooking-ingredient delivery products never did. Open-ended or ambitious tasks still go faster with Claude or ChatGPT.

  3. Worth knowing

    It works best when the answer is already in the context, so the effort goes into building a knowledge base: fridge contents, order history, and live stock data.

Example prompts

Try with Hermes Agent
Build me a grocery skill that reads my fridge list in [file or note], my past orders from [store], and current stock at [store website]. Then fill a cart with what I'm running low on for [number] days of meals. Pause and show me the cart and total before checkout, and stop once I've approved the order.
Content & Creative

Turn AI newsletters into a daily podcast

An agent runs overnight, collects the author's AI newsletter subscriptions and produces a daily podcast to listen to on walks.

Other or custom