MCP App Store
by KeenEthics
iPlugin Junqi icon

iPlugin Junqi MCP Server for ChatGPT

By Trustflow

Overview

Play a fullscreen face-up Luzhanqi (军棋) match against ChatGPT. You command 25 red pieces from the bottom; ChatGPT chooses every blue move from the complete 12 × 5 field and its legal move list. Move one step along a road, run a railway, turn corners with an engineer, use camps for safety, and capture the flag. Includes battle reports, legal-destination hints, complete-turn undo with fallen pieces restored, AI First, three difficulty levels, and confirmed resignation. No account, OAuth consent, separate model API, opening book, or hidden engine is used.

How to connect

  1. ChatGPT: Open the iPlugin Junqi listing in the ChatGPT app directory and choose Connect. ChatGPT listing iPlugin Junqi exposes 7 tools here.

Tools

junqi_ai_move

ChatGPT
Use this to apply the single legal blue move chosen by the current model from the complete board and the legalMoves list. No server engine, learning system, opening book or hidden selector is used.

junqi_get_state

ChatGPT
Use this to read the authoritative board, both armies, legal moves, move history, battle results and next turn for one junqi gameId. This tool only reads state.

junqi_home

ChatGPT
Use this only once to open the fullscreen Junqi (Luzhanqi) board for the conversation. It is the only UI-rendering tool and accepts no arguments. Never call it again after the board is open; use junqi_start_game in the existing view instead.

junqi_play_move

ChatGPT
Use this to apply one legal red move from the human. The server validates the route, camp and headquarters rules, resolves any battle, and returns the next state. Then choose the blue reply from the returned legalMoves.

junqi_resign

ChatGPT
Use this only after the human explicitly confirms resignation. It permanently ends the active game with the blue side as winner, and the game cannot be resumed.

junqi_start_game

ChatGPT
Use this when the human wants a new 1v1 face-up Junqi game. The human plays the 25 red pieces at the bottom; the model plays the 25 blue pieces at the top. Optionally choose who moves first, a difficulty level, and a setup seed for a reproducible deployment.

junqi_undo

ChatGPT
Use this during an active game to permanently remove the latest complete turn: the blue reply plus the red move, or a pending red move, or the blue opening. The prior position, including pieces fallen in the removed turn, is reconstructed; there is no redo.

Capabilities

InteractiveWrites

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/iplugin-junqi"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/iplugin-junqi)