MCP Catalogs
Home

KoikatuMCP

by great-majority·4·Score 33

MCP server that allows AI assistants to control Koikatu Studio through natural language commands.

ai-llmmediaother
0
Forks
0
Open issues
10 mo ago
Last commit
2d ago
Indexed

Overview

KoikatuMCP is a Model Context Protocol server that bridges AI assistants with Koikatu Studio, enabling control through natural language commands. It connects to the KKStudioSocket WebSocket plugin to manipulate scenes, objects, characters, and more. The server implements scene management, camera control, content discovery, and object manipulation features, making it possible for AI systems to interact with this character creation software.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Automating scene setup in Koikatu Studio through AI commands
you:Creating interactive experiences with AI-controlled character manipulation
you:Generating complex character scenes via natural language prompts
you:Which games does KoikatuMCP support?
you:How do I verify the connection between KoikatuMCP and Koikatu Studio?
you:Can I use KoikatuMCP with other AI assistants besides Claude?

When to choose this

Choose KoikatuMCP if you want to use AI to automate scene creation, character posing, or virtual photography in Koikatu Studio through natural language commands.

When NOT to choose this

Don't choose KoikatuMCP if you don't own or use Koikatu/Koikatsu Sunshine, or if you need integration with other 3D software beyond this specific game.

Tools this server exposes

12 tools extracted from the README
  • ping

    Test connection to Koikatsu Studio via KKStudioSocket

  • treetree(depth?: number)

    Retrieve scene hierarchy and structure of objects in the scene

  • add_object

    Add objects to the scene (items, lights, cameras, characters)

  • modify_object

    Modify object properties like position, rotation, scale, and color

  • remove_object

    Remove objects from the scene

  • camera_control

    Control camera position, orientation, and switch between views

  • capture_screenshot

    Capture screenshots of the current scene

  • browse_items

    Browse available item catalogs in Koikatsu Studio

  • search_items

    Search for specific items in the catalog

  • get_item_info

    Get detailed information about a specific item

  • attach_object

    Attach objects to create hierarchies

  • modify_lighting

    Adjust lighting properties in the scene

Comparable tools

ai-control-gaminggame-mcpvirtual-world-automationcharacter-creation-aiscene-design-ai

Installation

Installation

  1. Install KKStudioSocket Plugin

- Follow the [KKStudioSocket installation guide](https://github.com/great-majority/KKStudioSocket?tab=readme-ov-file#-installation) to set up the WebSocket plugin

  1. Download KoikatuMCP

- Download the latest release from [Releases](https://github.com/great-majority/KoikatuMCP/releases) - Extract the ZIP file to any location - Contains KoikatuMCP.exe (single executable file)

  1. Configure with Claude Desktop

- Add to your claude_desktop_config.json: ``json { "mcpServers": { "koikatu-mcp": { "command": "C:\\path\\to\\KoikatuMCP.exe", "env": { "KKSTUDIOSOCKET_URL": "ws://127.0.0.1:8765/ws" } } } } ``

  1. Start Koikatsu Studio with KKStudioSocket active, then start Claude Desktop to begin using.

FAQ

Which games does KoikatuMCP support?
KoikatuMCP supports both Koikatu (KK) and Koikatsu Sunshine (KKS) games, as long as the KKStudioSocket plugin is installed.
How do I verify the connection between KoikatuMCP and Koikatu Studio?
You can test the connection by asking your AI assistant to 'run a ping to check the connection' which will execute a `koikatu-mcp:ping` command to verify the connection to KKStudioSocket.
Can I use KoikatuMCP with other AI assistants besides Claude?
Yes, KoikatuMCP should work with any MCP client that can execute external commands. The provided configuration examples show how to set it up with Claude Desktop, but it may work with other AI systems that support MCP.

Compare KoikatuMCP with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.