beijing-fushengji-mcp
Enables AI agents to play a classic turn-based business game set in Beijing, autonomously making decisions like moving, trading, banking, and debt repayment over 40 days. Supports loading and saving game records for cross-session learning.
README
北京浮生记 MCP Server
经典回合制经商游戏「北京浮生记」的 MCP Server 实现。AI Agent 通过 MCP 工具自主完成整局 40 天的游戏——移动、买卖、存取款、还债,全程无需人类干预。

OpenClaw 的游戏
这是一款让 OpenClaw 自己玩的游戏。
OpenClaw 已经有了自己的社交平台 Moltbook,能在上面发帖、互动、交朋友。那 OpenClaw 当然也可以有自己的游戏——不是帮人类玩,是 AI 自己玩给你看。
你只需要说一句「来一局北京浮生记」,OpenClaw 就会化身北漂青年铁牛,独立完成 40 天的买卖、还债、存钱、跑路……全程自主决策,你只需要坐下来当观众。每隔 10 天铁牛还会给你"发朋友圈"汇报战况,游戏结束后还有一篇精彩的北漂故事回顾。
人类玩游戏,AI 看着。现在反过来了——AI 玩游戏,你看着。
游戏简介
你扮演北漂青年铁牛,带着 2000 元现金和 5000 元债务来到北京。在 40 天内通过倒卖 8 种商品(进口香烟、走私汽车、盗版 VCD 等)赚钱还债。债务每天涨 10% 复利,不还会膨胀到 22 万+。
最终得分 = 现金 + 银行存款 - 债务
提供的 MCP 工具
| 工具 | 说明 |
|---|---|
start_game |
开始新一局游戏 |
move |
移动到指定地点(消耗 1 天) |
buy / sell |
买入 / 卖出商品 |
bank |
银行存款或取款 |
repay_debt |
向村长还债 |
hospital |
去医院治疗 |
rent_house |
租更大的房子(扩容背包) |
get_status |
查看当前游戏状态 |
load_game_records |
加载历史游戏记录和经验总结 |
save_game_records |
保存游戏记录和经验总结 |
安装
前置条件
- Node.js >= 18
- npm
步骤
# 1. 克隆仓库
git clone <repo-url> beijing-fushengji-mcp
cd beijing-fushengji-mcp
# 2. 安装依赖
npm install
# 3. 编译 TypeScript
npm run build
编译后入口文件为 dist/index.js,通过 stdio 提供 MCP 服务。
在 OpenClaw 中安装
手动安装
完成上面的编译步骤后,执行以下三步将 Skill 注册到 OpenClaw:
# 1. 创建符号链接,让 OpenClaw 发现这个 Skill
ln -s "$(pwd)" ~/.openclaw/skills/beijing-fushengji-mcp
# 2. 注册 MCP Server 到 mcporter 配置
# 编辑 ~/.openclaw/workspace/config/mcporter.json,在 mcpServers 中添加:
# "beijing-fushengji": {
# "command": "node <绝对路径>/beijing-fushengji-mcp/dist/index.js"
# }
# 3. 重启 Gateway 生效
openclaw gateway restart
验证安装
# 检查 Skill 是否就绪
openclaw skills list | grep beijing
# 应显示: ✓ ready │ 📦 beijing-fushengji
Agent 自动安装指引
如果你是 AI Agent 并需要自动完成安装,按以下步骤执行:
- 将本项目克隆到
~/.openclaw/workspace/beijing-fushengji-mcp - 在项目目录执行
npm install && npm run build - 创建符号链接:
ln -s ~/.openclaw/workspace/beijing-fushengji-mcp ~/.openclaw/skills/beijing-fushengji-mcp - 读取
~/.openclaw/workspace/config/mcporter.json,在mcpServers对象中追加:注意:"beijing-fushengji": { "command": "node ~/.openclaw/workspace/beijing-fushengji-mcp/dist/index.js" }command中的路径必须展开为绝对路径(替换~) - 执行
openclaw gateway restart重启 Gateway - 执行
openclaw skills list | grep beijing确认状态为✓ ready
在其他 MCP 客户端中使用
在任何支持 MCP 的客户端(Claude Desktop、Cursor 等)的配置中添加:
{
"mcpServers": {
"beijing-fushengji": {
"command": "node",
"args": ["<绝对路径>/beijing-fushengji-mcp/dist/index.js"]
}
}
}
项目结构
src/
├── types.ts # 类型定义(GameState, GameResponse 等)
├── game-data.ts # 游戏数据(商品、事件、地点、常量)
├── game-engine.ts # 游戏引擎(状态管理、回合推进、事件触发)
└── index.ts # MCP Server 入口(注册 13 个工具)
records/
└── game_records.md # AI Agent 游戏记录与经验档案(跨局持久化,≤2000字符)
技术栈
- TypeScript 5.8(ES2022 / Node16 ESM)
@modelcontextprotocol/sdk— MCP 协议zod— 参数校验- 游戏状态纯内存,游戏记录持久化到
records/game_records.md
原版游戏
https://github.com/chrisguo/beijing_fushengji
许可证
GPL 2.0
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.