Compare dealership quotes and push for a lower price
An agent gathered offers from nearby dealerships and used them to negotiate a better car deal.
- What they did
- Aaron Stuyvenberg used MoltBot, a self-hosted agent that could handle a browser and email. He asked it to look up fair pricing on Reddit, contact 8 to 10 local dealerships, and collect quotes. It then completed dealer web forms and sent competing offers by email, while calls and texts were routed back to email.
- How it went
- After three days, he had a deal that cut more than $4,200 from the price. He only had to visit the showroom to sign the paperwork.
- Tips
- This kind of task works best when the agent can keep all contact in one place. Use a separate machine, plus a throwaway email and phone number, since email-based agents can be tricked by malicious messages.
Try it yourself with OpenClaw
Help me buy a [car make, model, trim] near [zip code]. First, look up fair market pricing and note the range you find. Then contact [number] dealerships that appear to have this car in stock, request an out-the-door quote, and keep a record of every offer. When a lower quote comes in, draft replies to the other dealers asking whether they can beat it, but stop and show me each form and email before anything is sent. If a dealer asks for a call or text, route it to my email instead. End with a ranked list of the quotes and the best total price.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
2 more reports of this
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Let your local model place a grocery order from live data
A local Qwen model used fridge inventory, old orders, and live shop stock to buy groceries online.
Check Shopify catalogs for daily product and stock updates
A custom agent watched Shopify product feeds each day and compared snapshots to find catalog, pricing, and inventory changes.
Set up an online shop and test the full checkout
They built a retail site with AI, added checkout, and verified the payment flow before turning it into a live store.
Use past orders to build your grocery cart
An Instacart assistant filled a cart from prior shopping history, then added meal ingredients and pointed out missing store items.
Automate weekly store updates from a spreadsheet
A custom agent keeps an online shop current by reading a weekly inventory sheet, revising listings, and asking for approval on unclear items.
Let an AI agent place a lunch order
A developer had an AI agent handle lunch ordering and payment, and the meal showed up without any more input.