BC Observatoryunofficial, made by waldo
community postCopilot & Agentscommunity - not Microsoftreviewed - checked by Opus

Copilot Inventory Queries

Copilot Inventory Queries is a free AppSource extension the author built to answer natural language inventory questions in Business Central, such as reorder points, replenishment timing and demand trends, with explanations. It sends the user question plus JSON inventory data like item ledger entries and open orders to an Azure OpenAI model, guided by a system prompt.

Open in Claude3585.md

post Olof Simren

Olof Simren · 2025-10-08 · 1,001 words · about 5 min

"The Copilot features and the underlying AI capabilities are what get people talking when we present Business Central to customers."
Show the post hereRead on Olof Simren

Key points

  • Built with the Copilot capability tools Microsoft provides for AL; the author found it simple to create
  • Needs an Azure OpenAI model from Azure AI Foundry; endpoint, deployment name and API key go in a setup table, with the API key kept in isolated storage
  • A system prompt tells the model it receives Business Central inventory data plus a question and must explain its answer
  • Item ledger entry data can be compressed and only relevant, configurable fields are sent to limit response time and cost
  • Example queries cover reorder point and quantity, typical sales volume and reordering policy; future ideas include MRP explanations and forecasting

Quotes

  • “The Copilot features and the underlying AI capabilities are what get people talking when we present Business Central to customers.” (It illustrates why Copilot is a key sales driver and customer interest point for Business Central in competitive positioning)
  • “It is surprisingly simple to create, you need to learn the Copilot interface features that Microsoft provides” (It shows that building custom Copilot features is accessible to partners when they understand the available tools and interfaces)

Context

  • Features: Copilot interface for inventory questions, Azure OpenAI backend integration, Inventory data analysis, Reorder point suggestions, Demand trend analysis, Automated explanations for AI recommendations

Source: Olof Simren, community blog. Summary, key points and quotes are derived (CONTENT-NOTICE.md); read the original for the full text.