What's New in Propamp.ai: Payment Calendar, Multi-Country Inventory & AI That Files Your Paperwork

Propamp.ai has rolled out a batch of platform updates aimed squarely at the problems Amazon sellers run into day to day. Most were built and battle-tested inside the team's own Amazon business before shipping. Here's a tour of what's new.

A payment calendar built for Amazon cash flow

Every seller keeps some version of a payment calendar — who to pay, when, and how much. The old spreadsheet approach is tedious and easy to get wrong.

Propamp.ai's new payment calendar lets sellers drag and drop payments onto any target date, and it automatically flags whether each payment is scheduled or overdue. Move a payment to a date that has already passed and it flips to overdue immediately.

That is especially useful on Amazon, where cash flow hinges on when disbursements land and cash gaps are common. When a payment needs to slip, it is a quick drag to a new day. Payments can be viewed by day, week, month, or across the full year, showing how much has been paid, how much is still due, and total spend — and each entry links back to the specific order it belongs to. It is the same view Propamp.ai's own finance team relies on to stay ahead of what is due.

MCP files the paperwork

The platform's MCP server connects an AI model directly to Propamp.ai, turning it into a set of hands for routine work.

From any purchase order, a seller can hand the AI its attachments — PDFs, Excel files, entry summaries, shipping documents — and ask it to file them against a specific order and into the right folder for the team. Feed it the details for a shipment and it handles the filing, so everyone knows where to look.

Multi-country inventory planning

Sellers working across many ASINs and multiple countries can store units in different locations. Done deliberately, that cuts delivery and storage costs and guards against stockouts, which are damaging to ranking and overall performance.

One example from the platform: on a single order, 736 units were purchased but only half shipped to Amazon. The remainder went to a third-party (3PL) warehouse to wait, since demand for that item is slower — with a draft shipment already prepared to move those units to the United States when they are needed. The purchase-order dashboard lets sellers filter by order and product to see the original quantity, the AI-recommended split across countries, and the residual quantity still held in the warehouse.

Stockout forecasts

The stock panel plots a snapshot of inventory on any given date, and the slope of the curve reflects sales velocity — raise the daily rate and the curve steepens accordingly.

Inventory held in a warehouse appears as a distinct band on the chart. When a draft shipment is scheduled to pull those units out, the forecast shows the band drawing down on that date as the units move into Amazon and begin selling — a clear picture of how a planned shipment prevents a stockout weeks ahead.

The forecast also projects long term. A model might show a stockout several months out, before seasonal demand — Black Friday, Cyber Monday, and the holidays — pulls that date forward. Toggling marketplaces on and off keeps the view accurate: inventory sitting in China or another country still counts as owned stock, even when it is destined for a different marketplace.

Amazon, AWD and 3PL in one view

Opening the inventory breakdown shows exactly how many units sit on Amazon, in AWD, in 3PL, and in total — for a single marketplace or across all of them at once.

For context, Amazon Warehousing & Distribution (AWD) lets sellers ship in bulk to an Amazon facility that then replenishes the fulfillment centers automatically, rather than shipping to FCs directly. 3PL covers whatever is held in a seller's own third-party warehouses. A recent touch to the performance chart also lets users zoom the full timeline or drag to focus on a specific window, making it easier to read performance over any period.

Bring your own AI

The MCP server can connect to a seller's own model or subscription — Anthropic, OpenAI, or others. Adding an API key routes usage through that existing plan (Claude, Codex, and similar) instead of spending platform tokens, with full visibility into token consumption. For sellers who already pay for a subscription, it avoids paying twice.

On the roadmap: workflow agents

One known limitation: a single prompt can consume a lot of tokens simply working out what is being asked, because it triggers many calls to the Amazon API. To address that, Propamp.ai is building dedicated agents for specific workflows — supply chain, Amazon PPC, seller support, and customer support — coordinated by a main agent that routes each request to the right specialist, cutting wasted tokens and cost.

Getting started

Propamp.ai offers free onboarding — sellers can connect an Amazon store and put these AI tools to work right away. Feedback from early users is welcome.