mcp-ready-mix

mcp-ready-mix

An MCP server that solves permission fatigue by letting you pre-approve commands, enabling Claude to run them without asking for permission each time.

Category
访问服务器

README

mcp-ready-mix

⚠️ Heavy Development Warning: This project is under active development. Use at your own risk. Breaking changes may occur without notice. Fix the tag you are working with by adding #vMajor.Minor.Patch to the end of the github repo name in your .mcp.json config.

A Model Context Protocol (MCP) server that solves the permission fatigue problem with Claude Code.

The Problem

When using Claude Code, you often need to run the same commands repeatedly (like tests, builds, or deployments). Claude asks for permission every single time, even for commands you've already approved. This quickly becomes tedious.

The Solution

Ready-mix lets you pre-approve a list of commands by adding them to a commands.json file. Claude can then run these commands without asking for permission each time.

Tools Provided

This server provides 5 tools, all validated with Zod schemas:

  • list_commands → lists all configured commands with their indices
  • run_command → executes a specific command by its index number
  • run_command_with_args → executes a command with arguments using placeholder substitution
  • get_version → returns the version from package.json
  • echo → echoes back a required text string

requirements

  • Node.js 18+
  • macOS/Linux (uses $SHELL, defaults to /bin/bash)

install

git clone https://github.com/yourname/mcp-ready-mix.git
cd mcp-ready-mix
npm install

local test

Single command:

echo '{"commands":[{"cmd":"ls -la","instructions":"List directory contents"}]}' > commands.json && MCP_COMMANDS_FILE="./commands.json" node server.mjs

From commands file:

MCP_COMMANDS_FILE="./commands.json" node server.mjs

This will print a startup line and wait for an MCP client.

For a quick smoke test of the echo logic:

node -e 'import("./server.mjs");'   # starts server, but you need an MCP client

use with claude code

1. Create your commands file

Create a commands.json file in your project root with this structure:

{
  "commands": [
    {
      "cmd": "npm test",
      "instructions": "Run all tests"
    },
    {
      "cmd": "npm run build",
      "instructions": "Build the project for production"
    },
    {
      "cmd": "npm run lint",
      "instructions": "Run linter to check code style"
    },
    {
      "cmd": "git status",
      "instructions": "Show git working tree status"
    },
    {
      "cmd": "git log --oneline -10",
      "instructions": "Show last 10 commits in compact format"
    },
    {
      "cmd": "python manage.py test",
      "instructions": "Run Django tests"
    },
    {
      "cmd": "git checkout {0}",
      "instructions": "Checkout git branch"
    },
    {
      "cmd": "npm test {0}",
      "instructions": "Run tests on specific file"
    },
    {
      "cmd": "grep -r {0} {1}",
      "instructions": "Search for pattern in directory"
    }
  ]
}

2. Add to your .mcp.json

Add this configuration to your .mcp.json file:

{
  "mcpServers": {
    "ready-mix": {
      "command": "npx",
      "args": ["-y", "github:Ivor/mcp-ready-mix", "mcp-ready-mix"],
      "env": {
        "MCP_COMMANDS_FILE": "/path/to/your/project/commands.json"
      }
    }
  }
}

Important: Replace /path/to/your/project/commands.json with the actual absolute path to your commands file.

The instructions field is optional but helpful for documenting what each command does.


Usage Examples

Adding/Updating Commands

To add new commands, simply edit your commands.json file:

{
  "commands": [
    {
      "cmd": "npm test",
      "instructions": "Run all tests"
    },
    {
      "cmd": "npm run build",
      "instructions": "Build the project"
    },
    {
      "cmd": "npm run lint",
      "instructions": "Check code style"
    },
    {
      "cmd": "git status",
      "instructions": "Show git status"
    },
    {
      "cmd": "python manage.py migrate",
      "instructions": "Run Django migrations"
    },
    {
      "cmd": "docker-compose up -d",
      "instructions": "Start services in background"
    }
  ]
}

Changes are picked up automatically - no restart needed.

Commands with Arguments

You can create commands that accept arguments using placeholder syntax {0}, {1}, etc:

{
  "commands": [
    {
      "cmd": "git checkout {0}",
      "instructions": "Checkout git branch"
    },
    {
      "cmd": "npm test {0}",
      "instructions": "Run tests on specific file"
    },
    {
      "cmd": "grep -r {0} {1}",
      "instructions": "Search for pattern in directory"
    },
    {
      "cmd": "find {0} -name {1}",
      "instructions": "Find files by name in directory"
    }
  ]
}

Use the run_command_with_args tool to execute these commands:

  • Command: git checkout {0} with args: ["main"] → executes git checkout 'main'
  • Command: grep -r {0} {1} with args: ["TODO", "src/"] → executes grep -r 'TODO' 'src/'

Arguments are automatically shell-escaped for security.

Example Claude Interactions

Once configured, you can ask Claude:

  • List the ready-mix commands → shows all available commands
  • Run command 1 → runs the first command (npm test)
  • Run command 3 → runs the third command (npm run lint)
  • Please run the build command → Claude will find and run your build command

For commands with arguments:

  • Run command 7 with args main → runs git checkout with "main" branch
  • Run command 8 with args tests/unit.test.js → runs npm test on specific file
  • Run command 9 with args TODO and src/ → searches for "TODO" in src/ directory

Typical Workflow

  1. Add commonly used commands to commands.json
  2. Ask Claude to run tests, builds, or other tasks
  3. Claude runs them instantly without asking for permission
  4. No more repetitive "Allow this command?" prompts!

Best Way to Test

The most effective way to test this tool is to install it in a real project and use the commands during actual development work. Set up a commands.json file with your most common development tasks (tests, builds, linting, git operations) and let Claude run them as you build your project. This gives you the true experience of permission-free command execution and helps identify which commands are most valuable for your workflow.


versioning

  1. bump version in package.json:
    npm version patch
    git push && git push --tags
    
  2. update .mcp.json args to the new tag or keep #main if you want latest.

license

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选