Skip to main content

BlueConic Model Context Protocol (MCP)

Updated over a week ago

BlueConic MCP enables AI tools to interact with your BlueConic tenant, facilitating exploration, retrieval, and analysis of your configuration and customer data. You can either run the BlueConic MCP client locally, or let your AI tool connect to your tenant MCP server.

Important: When you open your BlueConic tenant with MCP, you’re giving the model access to your environment. This means you should ensure you’re working with a trusted MCP host that handles data responsibly. If the host misuses the context, it could expose sensitive information. Always confirm that the MCP host does not use your data for training their models.

Key features:

  • Provides dynamic, read-only API access from your tenant’s OpenAPI spec.

  • Secured with OAuth2 tokens.

  • Works instantly with any BlueConic tenant.


Before you begin

Make sure you have the following:

  • An active BlueConic tenant.

  • OAuth2 credentials with the required read scopes.

  • An MCP-compatible tool such as Cursor, VS Code Copilot or Claude Desktop.


Create an OAuth2 Application in your BlueConic tenant

  1. Log in to your BlueConic tenant.

  2. Go to Settings > Access management > Applications.

  3. Select Add application.

  4. Create a new application with client credentials enabled.

  5. Grant the read scopes that you want the MCP extension to use.

  6. Copy the Client ID and Client Secret for use in your configuration.

    • The configured “run as user” permissions combine with the application scopes to determine access.

  7. Turn the application to ON.

  8. Click Save.


Connect to the public server

VSCode

In mcp.json, add the following configuration:
https://tenantname.blueconic.net/mcp points to the tenant's MCP endpoint.

  {
"servers": {
"blueconic": {
"name": "BlueConic MCP Server",
"description": "BlueConic MCP Server",
"type": "http",
"url": "https://tenantname.blueconic.net/mcp",
"headers": {
"X-BlueConic-Client-ID": "${input:blueconic-oauth2-client-id}",
"X-BlueConic-Client-Secret": "${input:blueconic-oauth2-client-secret}"
}
}
},
"inputs": [{
"type": "promptString",
"id": "blueconic-oauth2-client-id",
"description": "BlueConic OAuth2.0 Client ID",
"password": true
},{
"type": "promptString",
"id": "blueconic-oauth2-client-secret",
"description": "BlueConic OAuth2.0 Client secret",
"password": true
}]
}

Cursor

In mcp.json, add the following configuration:

{
"mcpServers": {
"blueconic": {
"name": "BlueConic MCP Server",
"description": "BlueConic MCP Server",
"type": "http",
"url": "https://tenantname.blueconic.net/mcp",
"headers": {
"X-BlueConic-Client-ID": "client-id-here",
"X-BlueConic-Client-Secret": "client-secret-here"
}
}
}
}

Or when using a token directly:

{
"servers": {
"blueconic": {
"name": "BlueConic MCP Server",
"description": "BlueConic MCP Server",
"type": "http",
"url": "https://tenantname.blueconic.net/mcp",
"headers": {
"Authorization": "${input:blueconic-oauth2-token}"
}
}
},
"inputs": [{
"type": "promptString",
"id": "blueconic-oauth2-api-token",
"description": "BlueConic Server OAuth2.0 Token",
"password": true
}]
}

Integration with AI Tools

Local MCP client

Cursor

Add the following to your .cursor/mcp.json:

{
"mcpServers": {
"blueconic": {
"command": "npx",
"args": ["@blueconic/blueconic-mcp"],
"env": {
"BLUECONIC_TENANT_URL": "https://yourtenant.blueconic.net",
"OAUTH_CLIENT_ID": "your_client_id_here",
"OAUTH_CLIENT_SECRET": "your_client_secret_here"
}
}
}
}

VS Code GitHub Copilot

Add to your VS Code settings mcp.json:

{
"servers": {
"blueconic": {
"name": "BlueConic MCP Server",
"description": "BlueConic MCP Server",
"command": "npx",
"args": ["@blueconic/blueconic-mcp"],
"env": {
"BLUECONIC_TENANT_URL": "${input:blueconic-tenant-url}",
"OAUTH_CLIENT_ID": "${input:blueconic-oauth2-client-id}",
"OAUTH_CLIENT_SECRET": "${input:blueconic-oauth2-client-secret}"
}
}
},
"inputs": [
{
"type": "promptString",
"id": "blueconic-tenant-url",
"description": "BlueConic tenant URL, e.g.
https://mytenant.blueconic.net",
"password": false
},
{
"type": "promptString",
"id": "blueconic-oauth2-client-id",
"description": "BlueConic OAuth2.0 Client ID",
"password": true
},
{
"type": "promptString",
"id": "blueconic-oauth2-client-secret",
"description": "BlueConic OAuth2.0 Client secret",
"password": true
}
]
}

Claude Desktop

Add the following to your claude_desktop_config.json:

{
"mcpServers": {
"BlueConic": {
"command": "npx",
"args": ["-y", "@blueconic/blueconic-mcp"],
"env": {
"BLUECONIC_TENANT_URL":
"https://your.sb.blueconic.net",
"OAUTH_CLIENT_ID": "xxxxxxxx",
"OAUTH_CLIENT_SECRET": "xxxxxx"
}
}
}
}
`

Other MCP tools

When your MCP tool doesn’t support sending two custom headers like demonstrated above, there is also a way to combine both the client-ID and client-secret into one header. This works for MCP tools like https://n8n.io/.

"X-BlueConic-Client-ID-Secret":"<client-ID>##<client-secret>"

Example prompts

Once configured, you can use natural language prompts to query your tenant:

  • “Get all segments from my BlueConic tenant and show me the commonalities.”

  • “Show me profiles in the ‘high-value-customers’ segment.”

  • “Retrieve 1000 profiles from the all visitors segment and summarize key insights.”

Example results


Next steps

  • Leverage MCP directly in your BlueConic environment using the AI Agent Connection to deliver insights directly to your inbox.

Did this answer your question?