Openakita/Skills@Mcp Installer logo

Openakita/Skills@Mcp Installer

CommunityPopular
openakita
openakita/skills@mcp-installer

Install, configure, and add MCP servers to the OpenAkita system. Use when the user needs to install MCP packages (npm/pip/uvx), connect remote HTTP/SSE MCP services, set up custom local MCP servers, or manage MCP server configuration and lifecycle.

Overview

Publisheropenakita
Repositoryopenakita
Skill nameopenakita/skills@mcp-installer
Stars
2K
Forks
277
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by openakita on GitHub. Read the source before you install it.

Installation

Install the Openakita/Skills@Mcp Installer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/openakita/openakita.git /tmp/openakita
mkdir -p .claude/skills
cp -r /tmp/openakita/skills/mcp-installer .claude/skills/openakita-openakita-skills-mcp-installer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Openakita/Skills@Mcp Installer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Openakita/Skills@Mcp Installer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Openakita/Skills@Mcp Installer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

MCP Installer — 安装与配置 MCP 服务器

系统 MCP 架构概述

OpenAkita 使用目录结构管理 MCP 服务器。每个 MCP 服务器是一个独立目录,包含配置和工具定义:

<server-name>/
├── SERVER_METADATA.json    # 必需:服务器配置
├── INSTRUCTIONS.md         # 可选:使用说明(复杂服务器建议提供)
└── tools/                  # 可选:工具定义(连接后可自动发现)
    ├── tool1.json
    └── tool2.json

配置存储位置

位置说明可写
mcps/内置 MCP(随项目发行)
.mcp/兼容目录
data/mcp/servers/用户/AI 添加的配置

所有新添加的 MCP 服务器写入 data/mcp/servers/

传输协议

协议场景必需字段
stdio本地进程(npx/python/node)command + args
streamable_http远程 HTTP 服务url
sse旧版 MCP 服务器(SSE)url

安装流程

方式一:使用 add_mcp_server 工具(推荐)

系统内置了 add_mcp_server 工具,可以直接添加 MCP 服务器:

stdio 模式(npx 包):

add_mcp_server(
    name="filesystem",
    transport="stdio",
    command="npx",
    args=["-y", "@anthropic/mcp-server-filesystem", "/path/to/dir"],
    description="文件系统访问"
)

stdio 模式(Python 包):

add_mcp_server(
    name="my-tool",
    transport="stdio",
    command="python",
    args=["-m", "my_mcp_package"],
    description="我的 MCP 工具",
    env={"API_KEY": "xxx"}
)

stdio 模式(uvx 包):

add_mcp_server(
    name="my-tool",
    transport="stdio",
    command="uvx",
    args=["my-mcp-package"],
    description="我的 MCP 工具"
)

streamable_http 模式(远程服务):

add_mcp_server(
    name="remote-api",
    transport="streamable_http",
    url="http://localhost:8080/mcp",
    description="远程 API 服务"
)

sse 模式(旧版兼容):

add_mcp_server(
    name="legacy-api",
    transport="sse",
    url="http://localhost:8080/sse",
    description="旧版 SSE 服务"
)

方式二:手动创建配置目录

直接在 data/mcp/servers/ 下创建目录结构。

第一步:创建目录

bash
mkdir -p data/mcp/servers/<server-name>

第二步:写入 SERVER_METADATA.json

json
{
  "serverIdentifier": "<server-name>",
  "serverName": "显示名称",
  "serverDescription": "服务器描述",
  "command": "npx",
  "args": ["-y", "package-name"],
  "env": {},
  "transport": "stdio",
  "url": "",
  "autoConnect": false
}

第三步(可选):创建 INSTRUCTIONS.md

为复杂的 MCP 服务器编写使用说明,Agent 可在需要时加载。

第四步(可选):预定义工具

tools/ 下创建工具定义 JSON(如果知道工具列表):

json
{
  "name": "tool_name",
  "description": "工具描述",
  "inputSchema": {
    "type": "object",
    "properties": {
      "param1": {
        "type": "string",
        "description": "参数描述"
      }
    },
    "required": ["param1"]
  }
}

工具定义是可选的——连接服务器后系统会自动发现工具。预定义工具的好处是在未连接时 Agent 也能在系统提示中看到工具列表。

第五步:加载配置

手动创建后调用 reload_mcp_servers 工具让系统扫描并加载新配置。


SERVER_METADATA.json 完整字段说明

字段类型必需说明
serverIdentifierstring唯一标识符,与目录名一致
serverNamestring显示名称
serverDescriptionstring简短描述
commandstringstdio 必需启动命令(python/npx/node/uvx 等)
argsstring[]命令参数
envobject环境变量
transportstring传输协议:stdio(默认)/streamable_http/sse
urlstringHTTP/SSE 必需服务 URL
autoConnectboolean启动时自动连接(默认 false)

兼容格式:"type": "streamableHttp" 等价于 "transport": "streamable_http"


常见 MCP 包安装示例

npm 包(通过 npx)

add_mcp_server(
    name="github",
    command="npx",
    args=["-y", "@modelcontextprotocol/server-github"],
    env={"GITHUB_PERSONAL_ACCESS_TOKEN": "<token>"},
    description="GitHub API"
)
add_mcp_server(
    name="puppeteer",
    command="npx",
    args=["-y", "@anthropic/mcp-server-puppeteer"],
    description="Puppeteer 浏览器自动化"
)
add_mcp_server(
    name="sqlite",
    command="npx",
    args=["-y", "@anthropic/mcp-server-sqlite", "path/to/db.sqlite"],
    description="SQLite 数据库"
)

Python 包(通过 python -m 或 uvx)

add_mcp_server(
    name="arxiv",
    command="uvx",
    args=["mcp-server-arxiv"],
    description="arXiv 论文搜索"
)
add_mcp_server(
    name="postgres",
    command="python",
    args=["-m", "mcp_server_postgres", "postgresql://user:pass@localhost/db"],
    description="PostgreSQL 数据库"
)

远程 HTTP 服务

add_mcp_server(
    name="composio",
    transport="streamable_http",
    url="https://mcp.composio.dev/partner/mcp_xxxx",
    description="Composio 集成平台"
)

本地创建的 MCP 服务器

如果使用 mcp-builder 技能创建了自定义 MCP 服务器,必须在创建后调用 add_mcp_server 注册:

Python 脚本(使用绝对路径):

add_mcp_server(
    name="my-custom-tool",
    command="python",
    args=["C:/path/to/my_project/server.py"],
    description="自定义 MCP 工具"
)

Python 模块:

add_mcp_server(
    name="my-custom-tool",
    command="python",
    args=["-m", "my_mcp_project.server"],
    description="自定义 MCP 工具"
)

TypeScript(编译后):

add_mcp_server(
    name="my-custom-tool",
    command="node",
    args=["C:/path/to/my_project/dist/index.js"],
    description="自定义 MCP 工具"
)

重要:本地脚本务必使用绝对路径,相对路径可能导致工作目录不对而失败。


安装前检查清单

  1. 确认命令可用:stdio 模式下检查 command 是否在 PATH 中(which npxwhich python
  2. 确认依赖已安装:npm 包需要 Node.js,Python 包需要对应环境
  3. 确认端口/URL 可达:HTTP/SSE 模式下确认目标 URL 可访问
  4. 准备环境变量:许多 MCP 服务器需要 API Key 等凭证,通过 env 字段传入
  5. 命名规范serverIdentifier 使用小写字母和连字符(如 my-tool),保持简洁

安装后验证

添加后系统会自动尝试连接。如果自动连接失败:

  1. 使用 connect_mcp_server("server-name") 手动连接
  2. 连接成功后使用 list_mcp_servers 查看状态
  3. 使用 call_mcp_tool("server-name", "tool_name", {...}) 测试调用

故障排查

问题可能原因解决方法
命令未找到未安装或不在 PATH安装对应运行时(Node.js/Python)
连接超时服务器启动慢或卡死增大 MCP_CONNECT_TIMEOUT(默认 60s)
HTTP 连接失败URL 错误或服务未启动确认 URL 正确且服务已运行
工具为空连接未成功先确保 connect_mcp_server 成功
权限错误API Key 缺失或无效检查 env 中的凭证配置

管理操作

  • 列出服务器: list_mcp_servers
  • 连接: connect_mcp_server("name")
  • 断开: disconnect_mcp_server("name")
  • 删除: remove_mcp_server("name")(仅 data/mcp/servers/ 中的配置)
  • 重新加载全部: reload_mcp_servers

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Openakita/Skills@Mcp Installer AI skill do?

Install, configure, and add MCP servers to the OpenAkita system. Use when the user needs to install MCP packages (npm/pip/uvx), connect remote HTTP/SSE MCP services, set up custom local MCP servers, or manage MCP server configuration and lifecycle.

Why use Openakita/Skills@Mcp Installer on TypingMind?

Because you install it once and use it with any model. Openakita/Skills@Mcp Installer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Openakita/Skills@Mcp Installer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/openakita/openakita/tree/main/skills/mcp-installer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Openakita/Skills@Mcp Installer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Openakita/Skills@Mcp Installer?

As many as you like. As long as a model supports skills, you can use Openakita/Skills@Mcp Installer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Openakita/Skills@Mcp Installer AI skill free?

Yes. It is published on GitHub by openakita under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇