A customer sends a photo with the question "how much will this cost", and the technician spends half an hour paging through TecDoc and supplier catalogues by VIN. The agent identifies the assembly and parts from the VIN and photo, finds their numbers in TecDoc, checks prices and lead times in the Inter Cars or LKQ portal and prepares an order that the technician approves.
Built for ISO 27001NIS2 DORA
Looking up parts in an auto repair shop is technician time taken away from repairs. You have to enter the VIN in TecDoc, pick the right variant from the list, check that the part really is the right one, then enter the same number in the Inter Cars or LKQ portal, compare prices and delivery times, and then call the customer with the total. When a customer sends a photo with the question "how much will this cost", the answer is often put off until evening because the technician is under the lift. A mistake in the variant means a returned part and downtime on the lift.
The agent receives the VIN and photo from the customer's message or the mechanic's phone. A Claude model with image recognition through the API identifies which assembly is in the photo and what damage is visible, and from the VIN, using the vehicle record in Tekmetric or Shopmonkey or a VIN decoder, confirms the variant, engine and trim. The agent then finds the required part numbers in the TecDoc catalogue, checks prices, stock and delivery times in the connected supplier portals, Inter Cars or LKQ, through an API or a login on your behalf, selects original and alternative options and builds a draft order: for each part the number, supplier, price, lead time and a link to the catalogue entry. The technician reviews the draft on a phone or computer, changes the supplier or option if needed and approves. Only after approval is the order sent to the supplier on your account, and the parts, with your markup, are added to that car's repair order in Tekmetric or Shopmonkey. The agent never spends money on its own and never builds orders without approval. We choose how to integrate the suppliers and the shop software, by API, export or login, after a check. The model runs on the Anthropic API or Amazon Bedrock in an EU region, photos and VINs are not used to train the model, and supplier logins are stored in your environment. The data protection on the workstations that technicians connect from is covered by our IT support for businesses.
Four steps from input to a record in your system, with human approval wherever there is risk.
The customer or mechanic sends a photo and a VIN, which the agent can read from a photo of the registration document or take from the car record in Tekmetric or Shopmonkey by number plate. The model identifies the photographed assembly and visible damage, and uses the VIN to confirm the variant and trim.
The agent finds the required part numbers in the TecDoc catalogue for the variant, then checks prices, stock and delivery times in the connected supplier portals, Inter Cars or LKQ, and builds a comparison table with original and alternative options according to the preference you set.
The technician reviews the table, picks the options and the supplier, and approves. The agent places the order in the supplier portal or by e-mail and adds the parts, with your markup, to that car's repair order or the customer's estimate in Tekmetric or Shopmonkey.
If the VIN leaves several possible variants in TecDoc, or none of the suppliers has the part in its catalogue, the agent says so and hands over to the technician with what it found. We maintain the supplier interfaces as they change and keep your preference rules up to date.
An agent that identifies the assembly from a photo and the variant in TecDoc from the VIN, and can work from a customer message, an e-mail or the shop software.
Your suppliers' portals, Inter Cars or LKQ, connected through an API or a login, with preference rules (original, alternative, lead time) and the markup logic.
A comparison table on a phone or computer where the technician picks parts and a supplier in one click, and the parts are written to the repair order in Tekmetric or Shopmonkey.
The history of every order with the photo, VIN, TecDoc numbers, selections and approval, and our support when supplier systems change.
Claude with image recognition through the API + the TecDoc catalogue and supplier portals (Inter Cars or LKQ) through an API or a login, parts into the Tekmetric or Shopmonkey repair order through an interface or export. We choose the platform by the task and by where your data may be located. Read more about this service and its limits on the service page.
No. An order in the Inter Cars or LKQ portal goes out only after the technician approves, and the agent has no payment methods. It prepares everything up to the approval button, so the technician only has to choose and click.
We use the TecDoc catalogue through the TecAlliance interface or the access you already have, and we connect supplier portals, Inter Cars or LKQ, through their APIs where you have one, or with a login on your behalf. We check each supplier's capabilities before implementation and choose the integration method after that check; we start with two or three main suppliers.
A photo helps identify which assembly is meant and what damage is visible, but part numbers are taken from TecDoc by VIN, not from the photo. When a photo is unclear, the agent asks for another or hands over to the technician instead of guessing. The first working solution usually takes 2–4 weeks.
An insurer form, an Audatex estimate, a manufacturer warranty claim, photos, parts invoices and the VIN: every case means an hour of paperwork and…
Mechanic notes on a scrap of paper, the service advisor's guesses and an invoice issued from memory in the evening.
Customers call over lunch, write on Messenger in the evening and leave voice messages when the service advisor is busy at the counter.
Tekmetric or Shopmonkey holds everything: last inspection, insurance date, tyre season and mileage, but nobody has time to remind customers.
This service is delivered by specialists with valid international certifications. Their expertise is confirmed by independent organisations, not by us.
AI systems are deployed with protection against data leakage and model abuse.
The AI security programme and its controls are managed following the ISACA methodology.
AI risks are identified and managed like any other IT risk.
We reply within 1 business day
