ELECTE allows you to expose your integrations and workspace data to AI tools through the Model Context Protocol (MCP).
MCP lets compatible AI tools securely interact with the data and integrations available in your ELECTE workspace. This can help you use your existing business data in AI-powered workflows without manually exporting or copying it.
Prerequisites
Before setting up MCP, make sure you have:
- An active ELECTE workspace
- The required permissions to manage integrations
- At least one connected data source or integration
- An MCP-compatible AI tool that you want to connect
Note: The AI tool you use must support the Model Context Protocol.
Set Up MCP
- Sign in to your ELECTE account.
- Click Integrations in the sidebar.
- Find MCP Server.
- Click Create API Key.
- Enter a name for the key.
- Click Create Key.
- Copy the API key and store it securely.
You will use this API key to authenticate your AI tool when connecting it to the ELECTE MCP Server.
Choose What to Expose
MCP allows AI tools to access the integrations and data available in your ELECTE workspace.
Before connecting an AI tool, review which data and integrations will be available through the MCP connection. Click Available Tools to see the available tools and their descriptions.
Use Your Data with AI Tools
After the MCP connection is configured, your compatible AI tool can interact with the integrations and data you have made available.
Manage Your MCP Connection
You can review or update your MCP configuration from Integrations → MCP Server.
If you no longer want an AI tool to access your ELECTE data, remove or disable the MCP connection.
Important: Disabling the MCP connection prevents the connected AI tool from accessing your ELECTE integrations through MCP. It does not delete the underlying data or integrations from your workspace.
Security Best Practices
- Expose only what you need. Avoid making unnecessary integrations or data available.
- Review access regularly. Remove MCP connections that are no longer required.
- Keep connection credentials secure. Do not share MCP connection details publicly.
- Check your workspace permissions. Make sure you understand which data the MCP connection can access.
Troubleshooting
If your AI tool cannot connect to the MCP server:
- Check that the MCP connection is enabled.
- Make sure the AI tool supports MCP.
- Check that you copied the connection details correctly.
- Confirm that the relevant integrations are connected and active.
- Review the permissions available to the MCP connection.