This is a high-level summary of the plugin’s core capabilities. For a more detailed list of currently available features you can have a look at the extended features list.

Feature Summary (At a Glance)

This is a high-level overview of the core functionalities.

• Universal Asset Architecture & Analysis: Generate, modify, refactor, and get detailed explanations for Blueprints, Materials, Behavior Trees, Animation Blueprints, and Widget Blueprints. Export any summary to disk for documentation.
• Project & Asset Management: Create new assets on the fly, including Structs, Enums, Blueprints, and folders, directly from a prompt.
• Real-Time Scene Direction: Spawn, select, and manipulate actors directly in the level viewport with conversational commands. (Experimental)
• Conversational UI Design: Design and prototype entire UMG layouts, create individual widgets, and modify their properties using natural language. (Experimental)
• Project-Wide Intelligence: Scan your entire project (Blueprints, Materials, etc.) to find performance hotspots and ask contextual questions about your codebase.
• External C++ Analysis: Read and write to external C++ source files (.cpp/.h) to analyze or modify external codebases.

Once you acquired the plugin you simply need to open up the Epic Games Launcher, access your Unreal Engine Library, look up for this plugin and install it to the engine.

Open up Unreal Engine and access the Plugins tab.

Enable the BP Generator and make sure these other plugins are enabled and restart the editor.

Restart the Editor.

That’s it, you can use the plugin right away but continue to read the following details:

The plugin by default has ~100k tokens free usage through an API key I’ve added. Once those ~100k tokens run out you’ll have to bring your own api key. You can use pretty much any provider and also you can connect the plugin to a local LLM if your machine is powerful enough.

Gemini offers a free api key. More details about Api Keys in the API KEYS section.

Version 0.3.5 released on 26th January fully integrates the plugin within the Unreal Editor without the need of using third party apps and connections like Claude Desktop, Cursor or connecting the plugin with the browser extension to DeepSeek.

The plugin can still be used as usual through Claude Desktop and other desktop apps like Cursor, Github Copilot, Google Antigravity etc. AND Local LLMs (with the 0.3.5 version).

GOOGLE GEMINI API KEY SETUP (free and recommended)

Video Overview

Head over to the Google Studio website so we can create a new API key.

Tap on the Create API key button.

Second option if you have no project created in google cloud already.

Copy the generated key and open back up your Unreal Engine project.

Add the key and Save.

OPEN AI API KEY SETUP (chatgpt)

While I recommend Gemini, OpenAI is a viable choice as well.

Head over to https://platform.openai.com/usage.

After logging in, tap on the Edit Budget text inside the Usage tab.

From the new window that has opened up, look for the Billing section.

Select Add to credit balance and you can go for the 5$ minimum credit. Once you add balance credit you can access the API keys section and create a new key.

Copy the generated key and head over to your Unreal Engine project.

Claude, DeepSeek API KEY works the same as OpenAi - you need to top up some credits on their website for the API key to work or you can have a valid subscription to them and you should be able to get a key.

This option is modular and will work with pretty much any api key provider. I’ve tested it with Z.ai and Venice.ai.

For the custom provider option you need to fill in the API endpoint, the model name and also the API Key. You’ll find these details on the provider’s website.

Ultimate Blueprint Generator supports local, OpenAI-compatible servers like Ollama and LM Studio. This allows for offline use, enhanced privacy, and no API costs.

IMPORTANT: The Endpoint URL

Your endpoint MUST end with /v1/chat/completions for the plugin to work correctly.

Do NOT use paths like /chat, /generate, or just the base URL. These will not work.

Ollama Setup

Step 1: Install & Run Ollama

1. Download from ollama.ai and install
2. Pull a model: ollama pull llama3.2
3. Run the model: ollama run llama3.2
4. Ollama automatically starts its API server in the background

Step 2: Configure in Plugin

In the plugin’s API Settings:

• Provider: Select Custom API Provider
• API Endpoint URL: http://localhost:11434/v1/chat/completions
• Model Name: In the main Blueprint Generator view, enter your model (e.g., llama3.2)

LM Studio Setup

Step 1: Start the Server

1. Open LM Studio and go to the Local Server tab (</> icon)
2. Select a model from the dropdown to load it
3. Click the Start Server button

Step 2: Configure in Plugin

In the plugin’s API Settings:

• Provider: Select Custom API Provider
• API Endpoint URL: http://localhost:1234/v1/chat/completions
• Model Name: In the main view, enter the exact model name shown in LM Studio’s dropdown

Troubleshooting

Connection issues?

• Make sure your local server is actually running before sending requests
• Verify the URL ends with /v1/chat/completions (NOT /chat or /generate)
• Check that the port matches (Ollama: 11434, LM Studio: 1234 by default)
• For networked machines, replace localhost with the IP (e.g., http://192.168.1.50:11434/v1/chat/completions)
• Check your firewall isn’t blocking the connection

Why /v1/chat/completions?

The plugin uses the OpenAI API format, which is the standard for local LLM servers. The /v1/chat/completions endpoint provides structured JSON responses that the plugin can parse correctly. Other endpoints like /chat or /generate use different formats and will not work.

Tested Local Models

• Ollama minimax m2

Download and install it from their Website: https://claude.ai/download

Once installed, run it and look for Settings:

In settings go to Extensions → Advanced Settings

Install Extension:

And go wherever the new version of the plugin is installed on your computer. (example: C:/Program Files/Epic Games/UE_5.7/Engine/Plugins/Marketplace/Ultimate3867945fbc2aV39/Content/Python/Extensions). And add them 1 by 1:

Once installed both you should see them running in the Developer tab.

Prime Claude by saying “how to use the connected tools?”

This is the default MCP config file but you need to change the path to point to wherever Epic installed the plugin:

{
  "mcpServers": {
    "unreal-handshake": {
      "command": "python",
      "args": [
        "C:/Program Files/Epic Games/UE_5.7/Engine/Plugins/Marketplace/Ultimate3867945fbc2aV39/Content/Python/mcp_main_server.py",
        "--server-name",
        "unreal-handshake"
      ],
      "env": {
        "UNREAL_HOST": "localhost",
        "UNREAL_PORT": "9877"
      }
    },
    "filesystem-access": {
      "command": "python",
      "args": [
        "C:/Program Files/Epic Games/UE_5.7/Engine/Plugins/Marketplace/Ultimate3867945fbc2aV39/Content/Python/mcp_main_server.py",
        "--server-name",
        "filesystem-access"
      ]
    }
  }
}

!! THE path you’ll copy has \ slashes instead of / slashes. Make sure to replace the \ with / !!

If you encounter difficulties make sure to consult an AI and it will be able to guide you further setting your app of choice.

You can customize the AI’s behavior by adding custom instructions that will be sent with each request.

Accessing Custom Instructions

Tap on the “Instructions” button inside the Plugin’s interface in the Unreal Editor.

Adding Your Instructions

In the newly opened tab, you’ll be able to append your custom instructions. These instructions are sent with every request to the AI, allowing you to modify the AI’s behavior to suit your specific needs.

Much more customization options will be added in future updates.

If you’re using the free Gemini API key, you may encounter certain error codes due to rate limiting.

Common Error Codes

You can check the official error codes documentation here: Gemini API Troubleshooting

Dealing with Rate Limits (Error 429)

If you encounter a 429 (Rate Limit) error:

• Create multiple free API keys using different Google accounts
• In a future update, an easy way to swap between API keys will be added
• Combine Gemini’s free tier with Claude Desktop’s free tier to minimize limitations
• Consider using a Local LLM for offline use without API costs

For more details or help, join the Discord Server.