Check wholesale order emails are complete before entry
Wholesale orders that arrive by email are checked by a decision model for the details needed to enter them, and incomplete orders get a drafted follow-up for staff to send instead of being entered with gaps.
An example pattern written by Agents Lookup. We have not yet found a public build of this exact job.
The use case
Wholesale buyers send orders as emails or attachments in their own formats, and orders entered with a missing ship-to address or purchase order number cause delays later. Jev reads each order email and answers whether each required detail is present. Orders where every detail is present with a confidence of 0.85 or higher are marked ready for entry. Orders with a missing detail, or low confidence, are marked incomplete, and Claude drafts a short follow-up asking for what is missing. Jev does not enter orders, set prices, or approve credit terms. Staff review each draft before sending it, enter complete orders, and decide on any pricing or credit question. Test it on a month of past orders, including the ones that caused problems, before relying on the ready list.
Example workflow
- Wholesale order emails land in an orders inbox in Gmail, and Make extracts the body and any attachment text.
- The text is sent to Jev as the state.
- Jev answers Noul questions: a purchase order number is present, a ship-to address is present, every line has a product code or clear name, every line has a quantity, and a requested ship date is present.
- Complete orders are added to an entry list in a Google Sheet.
- For incomplete orders, Claude drafts a follow-up listing the missing details, saved as a Gmail draft.
- Staff review and send the follow-ups and enter complete orders in Shopify or the order system.
Tools
- Jev (TypeSafe AI)
- Gmail
- Make
- Claude
- Google Sheets
- Shopify
Risks and guardrails
- A product name may be present but ambiguous — guardrail: staff match every line to a SKU when entering the order.
- Follow-ups sent automatically could confuse buyers — guardrail: drafts are saved only, and a person sends each one.
- Attachments in unusual formats may extract poorly — guardrail: orders with very little extracted text are marked for manual review.
Related recipe
Shopify Order-Exception Triage Agent: An agent reviews new Shopify orders for address problems, risk flags, stock gaps, split shipments, and customer notes, then sorts them, routes them, and drafts customer messages for a person to approve.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Gmail (official page): developers.google.com/workspace/gmail (checked 2026-11-19)
- Make MCP server (official page): developers.make.com/mcp-server (checked 2026-11-09)
- Claude (official page): claude.ai (checked 2026-11-09)
- Shopify Dev MCP server (AI Toolkit) (official page): shopify.dev/docs/apps/build/ai-toolkit (checked 2026-11-09)