View Categories

Connect an external AI (MCP)

3 minutes

Introduction #

The Model Context Protocol ( MCP ) allows you to connect an external AI client (Claude Desktop, Cursor, Windsurf, VS Code, JetBrains, etc.) directly to your Kafinea instance. The external AI can then query your ERP data, view records, and perform searches without leaving your usual work tool.

1. What is MCP? #

MCP is a standard protocol that allows AI clients to connect to tool servers. Kafinea provides an MCP server that grants secure access to ERP data, in accordance with the connected user’s access rights.

2. Go to the settings page #

  1. Sign in to Kafinea
  2. Go to My Preferences (user menu in the top-right corner)
  3. Click the MCP Configuration button (External AI)

The page displays your personal credentials and JSON configurations ready to copy for each supported AI client.

3. Your login credentials #

The configuration page displays three key pieces of information:

  • Access point (URL): the address of your Kafinea instance's MCP server
  • Username: your Kafinea username
  • Access Key: your personal access key

Important: Never share your access key. It grants access to your Kafinea data with your user permissions.

4. Set up your AI client #

For each AI client, the page provides a JSON configuration block that is ready to copy. Click the Copy button next to the desired client.

Claude Desktop #

  1. Open Claude Desktop
  2. Go to Settings → Developer → MCP Servers
  3. Paste the copied JSON configuration into the file claude_desktop_config.json
  4. Restart Claude Desktop

Cursor #

  1. Open Cursor
  2. Go to Settings → MCP → Add Server
  3. Select the JSON type
  4. Paste the copied configuration

Windsurfing #

  1. Open the file mcp_config.json Windsurfing
  2. Paste the copied configuration
  3. Restart Windsurf

VS Code / GitHub Copilot #

  1. Create or open the file .vscode/mcp.json in your project
  2. Or go to VS Code's user settings
  3. Paste the copied configuration

JetBrains (IntelliJ, PhpStorm, etc.) #

  1. Go to Settings → Tools → AI Assistant → MCP Servers
  2. Paste the copied configuration

5. Test the connection #

Once the configuration is in place:

  1. Restart your AI client if necessary
  2. Ask a question that refers to Kafinea data (for example: "List my recent invoices")
  3. The AI client should be able to query Kafinea and display the results

6. Safety #

  • Each user has their own access key
  • MCP requests adhere to the user's access rights
  • Never publish the JSON configuration in a public repository
  • If your access key has been compromised, generate a new one from My Preferences

FAQ #

My access key has changed. What should I do?
Go back to the MCP Configuration page and copy the configuration again using the new key.

The AI client cannot connect to
Verify that the access point URL is correct and accessible from your computer. Also verify that the MCP server has been enabled by your administrator.

What data can the external AI access?
Only the data you have access to in Kafinea, based on your profile and permissions.

Related references #