MCP server

Connect your AI assistant to beCPG PLM

Your AI assistant, connected to your products

Your teams already work with an AI assistant. It knows everything except what matters to them: your products.

beCPG changes that. Connect your teams’ assistant, such as Claude or Gemini, or the agent your IT department is building, to your PLM through MCP (Model Context Protocol), the open standard that links AI assistants to business software.

beCPG MCP server · tools
# beCPG MCP server: tools (excerpt)

# find
searchEntities
searchDocuments
listMyProjects
# read
getProductDetails
fetchProductHierarchy
queryDocumentContent
# analyse
compareProducts
validateProduct
runSavedQuery
runOptimization
# prepare (suggestions)
suggestCreateEntity
suggestUpdateEntityProperties
# regulatory
askRegulatoryQuestion
checkRegulatoryCompliance
In practice

What it changes day to day

A regulatory affairs managerasks her assistant to compare a customer specification received by e-mail with the formula in the PLM.
A product developerasks for the cost breakdown of three trials without opening the application.
A quality engineerhas a supplier specification sheet read and the raw material prepared, ready for validation.
Nobody needs to leave the toolthey already work in.
Tools

What the assistant can do

  • Find products, raw materials, packaging, projects and documents.
  • Read compositions, customer specifications, specification sheets and document content.
  • Analyze: compare products, check a product, run your saved business intelligence queries, optimize a recipe’s cost.
  • Prepare products, raw materials or projects as suggestions.
  • Ask regulatory questions and check a product’s compliance.
Security

Under your control

  • Your access rights: the assistant works with the signed-in person’s rights. It sees what they see, nothing more.
  • Your usual account: each user signs in with OAuth, using their beCPG account.
  • Your rules: changes go through suggestions, applied according to the policy set for your instance.
  • The beCPG engines still perform the regulatory and nutritional calculations: results remain reproducible.
Deployment

Setting it up

beCPG activates the MCP server on your instance. Your IT department adds its address to the assistant; each user then signs in with their own account.

The MCP server is part of the beCPG AI module. Configurations are provided for Claude, Gemini, Cursor and Visual Studio Code.

ClaudeGeminiCursorVisual Studio Code

Want to see it with your own assistant? We connect it during the demo.

Request a demo
What is MCP?

The Model Context Protocol is an open standard that lets an AI assistant use the tools of a business application: search, read, calculate, propose.

Which assistants can be connected?

Configurations are provided for Claude, Gemini, Cursor and Visual Studio Code. For another assistant or your own agent, we check compatibility with you.

Does the assistant see the whole PLM?

No. It works with the signed-in user’s access rights and only sees what that user can see.

Do we need Chef AI?

The MCP server is part of the beCPG AI module. Chef AI and the MCP server use the same tools and the same access rights.

Connect your assistant during the demo

Bring your own assistant: we connect it to a beCPG instance and you ask your questions.