原型 · 预览版:收录官方注册中心里的全部服务器,每 30 分钟同步一次。评价和账户功能将在后续阶段推出。
PopNx MCP为使用 AI 的人准备的 MCP 服务器 · 用大白话讲清楚

Funplay Cocos MCP

作者:FunplayAI版本 0.7.0i★ –i在你的电脑上运行
注册名称io.github.FunplayAI/funplay-cocos-mcp

安装

概览

作者提供的说明(英文原文)

stdio bridge for the local Cocos Creator Editor MCP server from FunplayAI/funplay-cocos-mcp.

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

  • 连接方式: 在你的电脑上运行
  • 传输方式: stdio
  • 分类: 其他

安装

选择你使用的 AI 应用,然后复制设置。

安装之前

  • 基础检查:1 项中通过 1 项
  • 在你的电脑上运行,拥有你的用户账户的权限(文件、网络、应用)。
  • 需要你自己填写: FUNPLAY_COCOS_MCP_URL, FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS

你需要

  • npm已安装 Node.js(用于 npx 命令)

你用的是哪个应用?

在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。

打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json

{
  "mcpServers": {
    "funplay-cocos": {
      "command": "npx",
      "args": [
        "-y",
        "funplay-cocos-mcp"
      ],
      "env": {
        "FUNPLAY_COCOS_MCP_URL": "your-value",
        "FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS": "your-value"
      }
    }
  }
}

在终端中运行这条命令

claude mcp add funplay-cocos --env FUNPLAY_COCOS_MCP_URL=your-value --env FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS=your-value -- npx -y funplay-cocos-mcp

运行前请先填写空着的值。

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "funplay-cocos": {
      "command": "npx",
      "args": [
        "-y",
        "funplay-cocos-mcp"
      ],
      "env": {
        "FUNPLAY_COCOS_MCP_URL": "your-value",
        "FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "funplay-cocos": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "funplay-cocos-mcp"
      ],
      "env": {
        "FUNPLAY_COCOS_MCP_URL": "your-value",
        "FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS": "your-value"
      }
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。

你的选择通过 Cookie 保存在这台设备上,无需账户。

需要你填写的值

  • FUNPLAY_COCOS_MCP_URL env
    Optional local Cocos MCP HTTP endpoint. Defaults to http://127.0.0.1:8765/.
  • FUNPLAY_COCOS_MCP_TIMEOUT_SECONDS env
    Optional HTTP timeout per request in seconds. Defaults to 120.

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。

它能做什么

它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。

基础检查i

我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月9日

  • ✓ 通过

    作者与仓库一致

    注册名称已通过 GitHub 账户 FunplayAI 验证,该账户拥有此仓库

  • … 尚未检查

    仍在维护

    我们还没有从 GitHub 读取这个仓库

  • … 尚未检查

    有许可证

    我们还没有从 GitHub 读取这个仓库

  • – 不适用

    端点有响应

    它在你的电脑上运行,没有互联网端点

  • – 不适用

    使用 OAuth 登录

    在你的电脑上运行,改用本地设置

  • … 尚未检查

    工具描述中没有隐藏指令

    它在你的电脑上运行;我们只能读取互联网端点的工具列表

与“其他”中的其他服务器比较i

  • 星标数多于此分类中其他 18665 个服务器的 0%
  • 发布到注册中心的时间比此分类中其他 18665 个服务器的 78% 更新
  • 基础检查结果等于或优于此分类中其他 18665 个服务器的 100%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。

常见问题

Funplay Cocos MCP 安全吗?

它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。

它需要哪些密钥或代码?

注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。

它在我的电脑上运行还是在云端?

它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。

它还在维护吗?

GitHub 仓库最后更新于–,最新版本(0.7.0)于11天前进入注册中心。

哪些应用可以使用它?

支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。

有什么可以替代它?

同一分类中有:Spec Driven Development, Remote Desktop Commander, Systemdox

替代选择

同一分类中其他作者的服务器,按星标数排序。

如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Spec Driven Development

MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.

作者:formulahendry起步阶段↓ 66/周★ 44014个月前更新

在你的电脑上运行

Remote Desktop Commander

Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.

作者:desktop-commander正在增长★ 2114天前更新

在云端运行✓ desktopcommander.app

Systemdox

Your architecture in your agents' context: checks, specs, ADRs and docs over MCP, at write time.

作者:puglieseweb正在增长↓ 3028/周今天更新

在你的电脑上运行

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

Funplay MCP for Cocos

An Embedded MCP Server for Cocos Creator Editor

中文 | English

If this project helps your Cocos workflow, please consider giving it a Star. It helps more developers discover the project and supports ongoing development.


Funplay MCP for Cocos is an MIT-licensed Cocos Creator extension that embeds an HTTP MCP server directly inside the editor. It lets AI assistants such as Claude Code, Cursor, Codex, VS Code Copilot, Trae, Kiro, Qoder, Kimi Code, OpenCode, CodeBuddy, and DeepSeek Harness (DSH) inspect and operate your running Cocos project.

The package follows the same product direction as Funplay MCP for Unity: keep the default tool surface focused, provide a one-click client configuration window, and make one high-flexibility execution tool the primary workflow.

For Cocos, the primary tool is execute_javascript:

  • context: "scene" runs JavaScript in the active Cocos scene/runtime context
  • context: "editor" runs JavaScript in the Cocos editor/browser context

*"Create a login page UI with account/password fields and a main login button."*

Your AI assistant can call execute_javascript, build the UI hierarchy under the active Canvas, attach Cocos components, inspect the result, and capture a screenshot for validation.

Quick Start

If you just want to connect quickly, do these three things:

  • Install this repository as a Cocos Creator extension
  • Open Funplay > MCP Server
  • Use the built-in one-click MCP client configuration
1. Install as a Cocos Creator Extension

Recommended install: use the official Cocos Store page and install the extension into your target Cocos Creator project.

For local development or unreleased builds, clone or copy this repository into your Cocos project extension directory:

cd /path/to/your-cocos-project

mkdir -p extensions git clone https://github.com/FunplayAI/funplay-cocos-mcp.git extensions/funplay-cocos-mcp

Then restart Cocos Creator or reload extensions from the editor.

For a non-git install, download Funplay.CocosMcp.v .zip from the GitHub Releases page, unzip it, and move the extracted funplay-cocos-mcp folder into your project extensions/ directory.

Install for All Projects (Recommended)

Install and open the extension in any project first. Then open Funplay > MCP Settings and click Install for All Projects in the All Projects section. The extension downloads the latest GitHub Release, verifies SHA256SUMS.txt, and installs it into the active Cocos Creator version's managed global extension directory:

  • macOS / Linux: ~/.CocosCreator/builtin-extensions/ /funplay-cocos-mcp
  • Windows: %USERPROFILE%\.CocosCreator\builtin-extensions\ \funplay-cocos-mcp

Cocos Creator scans and registers the installed copy immediately. Projects opened or created with that Creator version then load it automatically, and the MCP server keeps its default autostart behavior. Reopen an already-running project if its project-level copy has priority. No extension files are copied into each project's extensions/ directory.

Global extensions are isolated by Cocos Creator version. If you use multiple Creator versions, run Install for All Projects once from each version.

The installer does not delete the current project copy, which protects development worktrees. If both copies exist, MCP Settings shows a duplicate-install warning. After confirming that the global copy works, remove or disable the project copy to avoid extension conflicts.

2. Start the MCP Server

Open the editor menu:

Funplay > MCP Server

New projects get a stable project-derived port in 20000–29999. Existing saved configurations keep their fixed port (including 8765). Use the exact URL shown in MCP Server; if a temporary fallback port is active, click Pin Current Port or Use Per-Project Port before configuring a client.

If the configured port is already occupied, the extension first checks whether the existing listener belongs to the same Cocos project. Same-project listeners are reused safely; unrelated listeners trigger automatic fallback to the next available local port.

The panel is intentionally small:

  • Enable or disable the MCP server
  • Open focused Tool Exposure, MCP Settings, and Project Skills windows
  • Automatically check the installed version against the latest GitHub release
  • Open the release page or install verified release packages from the panel
  • Install once in the Cocos Creator global extensions directory so existing and new projects can use it automatically
  • Configure AI clients with one click and preview the selected client config
  • Keep advanced tool profile editing, transport settings, and installation management out of the main server window
  • Follow the Cocos Creator interface language automatically, or select Chinese/English per project in MCP Settings
  • Toggle Print MCP logs in Settings to silence MCP console messages without losing Recent Activity or internal diagnostics

All management pages use consistent headings, aligned controls, and lightweight sections. Tool Exposure groups per-tool toggles by category; Project Skills highlights available actions and folds away file paths and backups. Recent Activity shows each call's summary and actual result as an expandable object/array tree, with execution logs and return values in separate sections for scripts. Output sections and the standalone log-viewing panel are removed. Short action notices preserve success/failure feedback, while diagnostic logs remain available to MCP tools and resources.

Script tools capture per-call console.log/info/warn/error/debug output, including logs before a failure, independently of Print MCP logs. Capture is isolated across concurrent and nested calls and stops when the script settles; global/project logs and logs from separately required modules are not collected. Use the injected console (also available as env.console in run/module exports). MCP responses retain their existing data shape and add optional execution metadata with context, duration, bounded/redacted logs, and an omitted-entry count. Recent Activity preserves result hierarchy and scalar values; large results and file/script bodies remain limited or omitted, with truncation notices.

The extension menu and native panel titles follow the Cocos Creator interface language. Panel controls, tool descriptions, status messages, prompts, and empty states support English and Chinese; open Funplay > MCP Settings to follow Creator or set a project-specific content-language override. MCP tool IDs and schemas remain in English so client integrations stay stable.

3. Configure Your AI Client

Use the built-in MCP Client Config section in the Funplay > MCP Server panel first.

Select your target client, click One-Click Configure, and the extension writes the recommended MCP config entry for you.

The MCP server name written to clients is:

cocos-<project>-<hash>

If you prefer to edit config files manually, use the examples below as fallback references. Replace their legacy funplay_cocos name and 8765 port with the name and URL shown in your panel. One-click Claude Code configuration uses the project's Git-root scope in ~/.claude.json, not the global example below. See project setup and migration.

Claude Code / Claude Desktop

{

"mcpServers": { "funplay_cocos": { "type": "http", "url": "http://127.0.0.1:8765/" } } }

Cursor

{

"mcpServers": { "funplay_cocos": { "url": "http://127.0.0.1:8765/" } } }

VS Code

{

"servers": { "funplay_cocos": { "type": "http", "url": "http://127.0.0.1:8765/" } } }

Trae

{

"mcpServers": { "funplay_cocos": { "url": "http://127.0.0.1:8765/" } } }

Kiro

{

"mcpServers": { "funplay_cocos": { "type": "http", "url": "http://127.0.0.1:8765/" } } }

Qoder

The user-level configuration is written to ~/.qoder/settings.json, or to $QODER_CONFIG_DIR/settings.json when that environment variable is set.

{

"mcpServers": { "funplay_cocos": { "type": "http", "url": "http://127.0.0.1:8765/" } } }

Kimi Code

The user-level configuration is written to ~/.kimi-code/mcp.json, or to $KIMI_CODE_HOME/mcp.json when that environment variable is set.

{

"mcpServers": { "funplay_cocos": { "url": "http://127.0.0.1:8765/" } } }

Codex

[mcp_servers.funplay_cocos]

url = "http://127.0.0.1:8765/"

OpenCode

The global config lives in $XDG_CONFIG_HOME/opencode/opencode.json or ~/.config/opencode/opencode.json on every platform (opencode.jsonc also works).

One-click configuration prefers an existing opencode.jsonc. It supports comments and trailing commas in either file, preserves unrelated settings and comments, and leaves malformed or ambiguous configurations unchanged.

{

"mcp": { "funplay_cocos": { "type": "remote", "url": "http://127.0.0.1:8765/" } } }

CodeBuddy (IDE 4.0+)

One-click setup writes /.mcp.json, the project configuration supported by CodeBuddy IDE 4.0+. It preserves other servers, comments, and trailing commas. Open this Cocos project folder as the CodeBuddy workspace, reload the client if necessary, and enable/approve the MCP server when prompted. Older IDE versions and the standalone CodeBuddy editor extension have different configuration locations; this target does not write their global settings or the CLI's global files.

Configure + Skills also installs missing built-in Skills into /.codebuddy/skills, following the official Skills directory. Existing Skills are kept; review updates in Project Skills.

{

"mcpServers": { "funplay_cocos": { "type": "http", "url": "http://127.0.0.1:8765/" } } }

DeepSeek Harness (DSH)

Select DeepSeek Harness (DSH) in the MCP Server panel. Configure generates /.dsh/funplay-cocos-mcp.patch.json, a JSON-formatted YAML overlay for the official Cordis MCP client. It does not change $DSH_HOME, global/profile YAML, dependencies, credentials, or approval settings.

The overlay is not auto-loaded. Start DSH from the Cocos project directory using the exact command shown in the panel, or add its `--

只显示了开头部分;其余内容请在仓库中阅读

问 PopNx