Export

Stream your Supernova prototypes and documents directly into AI coding assistants.

Plan: All

What is Export?

Export allows you to stream prototype and document data to MCP server or download generated code.

Export

How it works

From the Export page, you can start streaming directly to your MCP server. Stream a prototype to publish the latest code iteration, or stream a document to keep your knowledge base fully up to date.

When you switch to a different prototype or document in Supernova, the stream updates automatically. You can stream one prototype and one document at the same time, and all streams remain securely tied to your account.


Enabling the MCP server

Role: Builder+

The Supernova MCP server connects AI coding tools like Cursor and Claude Code directly to your Supernova projects. Instead of copying code manually, AI pulls components, pages, and themes from a selected feature, plus context from related documents.


Cursor

1. Click the Supernova MCP server deep link to open the MCP configuration in Cursor.

2. Click Connect to authenticate with Supernova.

3. Log in to Supernova and open Cursor.

Start prompting!


VS Code

1. In VS Code, use the shortcut ⌘ + Shift + P, and run MCP: Add Server.

2. Choose HTTP as the type of MCP server.

3. Paste the following URL into the Enter Server URL field: https://mcp.supernova.io/mcp

4. Enter Server ID: use the default ID or create a custom one.

5. Select the configuration target:

  • Global: to use Supernova MCP sever globally.
  • Workspace: to use Supernova MCP server only in your current workspace.

6. Click Allow and authenticate with Supernova.

Start prompting!


Claude Code

1. Open your terminal or IDE with Claude Code installed.

2. Install Supernova MCP by running following command: claude mcp add --transport http supernova-mcp 'https://mcp.supernova.io/mcp'

3. Start Claude Code by running following command: claude

4. Open MCP servers list by running following command: /mcp

5. Select supernova-mcp from the list and authenticate with Supernova.

Start prompting!


Windsurf

1. In Windsurf, open Settings / Windsurf Settings.

2. Go to to Cascade section and click Open MCP Marketplace.

3. In MCP Marketplace, click on settings icon in the top right corner to open the mcp_config.json file.

4. Paste the following code in mcp_config.json file:


                                                        
                                                        
                                                            {
                                                          "mcpServers": {
                                                            "mcp-supernova": {
                                                              "serverUrl": "https://mcp.supernova.io/mcp"
                                                            }
                                                          }
                                                        }
                                                        
                                                            

5. Save the mcp_config.json file, and authenticate with Supernova.

Start prompting!


ChatGPT

1. In ChatGPT, open Settings.

2. Go to Apps & Connectors.

3. Open Advanced settings and enable Developer mode.

4. Return to Apps & Connectors and click Create.

5. Paste the following to MCP Server URL: https://mcp.supernova.io/mcp

6. Click Create and authenticate with Supernova.

Start prompting!


Claude

1. In Claude, open Settings.

2. Go to Connectors and click Add custom connector.

3. Paste the following URL into the Remote MCP server URL field: https://mcp.supernova.io/mcp

4. Click Add to add this connector.

5. Click Connect and authenticate with Supernova.

Start prompting!


Tool function list

The following functions are currently available. They come with sufficient descriptions such that the tools should automatically pick up when to use them without any additional prompting needed.

get_me

Checks which account you are logged in to.

get_project_feature_artifact_details

Fetches a specific feature artifact by its ID only. Returns artifact id, name and full file content.

get_selected_project_document_details

Fetches selected document details and its full content in Markdown format.

get_selected_project_feature_details

Fetches selected feature details and its latest iteration artifacts.