[{"_1":2,"_317":-5,"_318":-5},"loaderData",{"_3":4,"_22":23,"_24":25,"_258":259},"root",{"_5":6,"_7":8},"toasts",[],"visitor",{"_9":10,"_11":12},"viewportWidth",1280,"visitorSession",{"_13":14,"_16":17,"_18":-5,"_19":20,"_21":-5},"attributes",[15],"BOT","id",7058884,"membership","referenceId","01971406-43e3-74e8-b880-2f000258007a","userAccount","routes/_public/_layout",{},"routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_26":27,"_68":-5,"_69":28,"_29":70,"_175":176,"_257":145},"availableLocales",[28,40,47,54,61],{"_29":30,"_33":37,"_38":39},"mcpServer",{"_31":32,"_33":34,"_35":36},"description","Allows AI assistants to safely execute common Unix/macOS terminal commands through a controlled interface with comprehensive security features.","name","Command-Line MCP Server","readme","# Command-Line MCP Server\n\n[![PyPI version](https://badge.fury.io/py/cmd-line-mcp.svg)](https://badge.fury.io/py/cmd-line-mcp)\n[![Python Versions](https://img.shields.io/pypi/pyversions/cmd-line-mcp.svg)](https://pypi.org/project/cmd-line-mcp/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\nA secure Model Control Protocol (MCP) server that allows AI assistants to execute terminal commands with controlled directory access and command permissions.\n\n## Overview\n\nCommand-Line MCP provides a security layer between AI assistants and your terminal. It implements a dual security model:\n\n1. **Command Permissions**: Commands are categorized as read (safe), write (changes data), or system (affects system state), with different approval requirements\n2. **Directory Permissions**: Commands can only access explicitly whitelisted directories or directories approved during a session\n\nAI assistants interact with this server using standardized MCP tools, enabling safe terminal command execution while preventing access to sensitive files or dangerous operations. You can configure the security level from highly restrictive to more permissive based on your needs.\n\n## Key Features\n\n| Security | Usability | Integration |\n|----------|-----------|-------------|\n| Directory whitelisting | Command categorization (read/write/system) | Claude Desktop compatibility |\n| Command filtering | Persistent session permissions | Standard MCP protocol |\n| Pattern matching | Command chaining (pipes, etc.) | Auto-approval options |\n| Dangerous command blocking | Intuitive approval workflow | Multiple config methods |\n\n## Supported Commands (out of the box)\n\n### Read Commands\n- `ls`, `pwd`, `cat`, `less`, `head`, `tail`, `grep`, `find`, `which`, `du`, `df`, `file`, `sort`, etc.\n\n### Write Commands \n- `cp`, `mv`, `rm`, `mkdir`, `rmdir`, `touch`, `chmod`, `chown`, etc.\n\n### System Commands\n- `ps`, `top`, `htop`, `who`, `netstat`, `ifconfig`, `ping`, etc.\n\n## Security Architecture\n\nThe system implements a multi-layered security approach:\n\n```\n┌───────────────────────────────────────────────────────────────┐\n│ COMMAND-LINE MCP SERVER │\n├──────────────────┬────────────────────────┬───────────────────┤\n│ COMMAND SECURITY │ DIRECTORY SECURITY │ SESSION SECURITY │\n├──────────────────┼────────────────────────┼───────────────────┤\n│ ✓ Read commands │ ✓ Directory whitelist │ ✓ Session IDs │\n│ ✓ Write commands │ ✓ Runtime approvals │ ✓ Persistent │\n│ ✓ System commands│ ✓ Path validation │ permissions │\n│ ✓ Blocked list │ ✓ Home dir expansion │ ✓ Auto timeouts │\n│ ✓ Pattern filters│ ✓ Subdirectory check │ ✓ Desktop mode │\n└──────────────────┴────────────────────────┴───────────────────┘\n```\n\nAll security features can be configured from restrictive to permissive based on your threat model and convenience requirements.\n\n## Quick Start\n\n```bash\n# Install\ngit clone https://github.com/yourusername/cmd-line-mcp.git\ncd cmd-line-mcp\npython -m venv venv\nsource venv/bin/activate\npip install -e .\ncp config.json.example config.json\n\n# Run\ncmd-line-mcp # With default config\ncmd-line-mcp --config config.json # With specific config\n```\n\n### Configuration Options\n\nThe server supports four configuration methods in order of precedence:\n\n1. **Built-in default configuration** (default_config.json)\n2. **JSON configuration file** (recommended for customization)\n ```bash\n cmd-line-mcp --config config.json\n ```\n3. **Environment variables** (for specific overrides)\n ```bash\n export CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"~,/tmp\"\n ```\n4. **.env file** (for environment-specific settings)\n ```bash\n cmd-line-mcp --config config.json --env .env\n ```\n\nThe default configuration is stored in `default_config.json` and is included with the package. You can copy this file to create your own custom configuration.\n\n#### Core Configuration Settings\n\n```json\n{\n \"security\": {\n \"whitelisted_directories\": [\"/home\", \"/tmp\", \"~\"],\n \"auto_approve_directories_in_desktop_mode\": false, \n \"require_session_id\": false,\n \"allow_command_separators\": true\n },\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\"], \n \"write\": [\"touch\", \"mkdir\", \"rm\"],\n \"system\": [\"ps\", \"ping\"]\n }\n}\n```\n\n#### Environment Variable Format\n\nEnvironment variables use a predictable naming pattern:\n```\nCMD_LINE_MCP_

_\n```\n\nExamples:\n```bash\n# Security settings\nexport CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"/projects,/var/data\"\nexport CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE=true\n\n# Command additions (these merge with defaults)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq,wc\"\n```\n\n### Claude Desktop Integration\n\n#### Setup\n\n1. Install [Claude for Desktop](https://claude.ai/download)\n2. Configure in `~/Library/Application Support/Claude/claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"cmd-line\": {\n \"command\": \"/path/to/venv/bin/cmd-line-mcp\",\n \"args\": [\"--config\", \"/path/to/config.json\"],\n \"env\": {\n \"CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID\": \"false\",\n \"CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE\": \"true\"\n }\n }\n }\n}\n```\n\n#### Recommended Claude Desktop Settings\n\nFor best experience, configure:\n- `require_session_id: false` - Essential to prevent approval loops\n- `auto_approve_directories_in_desktop_mode: true` - Optional for convenient access\n- Include common directories in your whitelist\n\nAfter configuration, restart Claude for Desktop.\n\n## AI Assistant Tools\n\nThe server provides these MCP tools for AI assistants:\n\n| Tool | Purpose | Needs Approval |\n|------|---------|----------------|\n| `execute_command` | Run any command type | Yes, for write/system commands |\n| `execute_read_command` | Run read-only commands | Directory approval only |\n| `approve_directory` | Grant access to a directory | N/A - it's an approval tool |\n| `approve_command_type` | Grant permission for command category | N/A - it's an approval tool |\n| `list_directories` | Show authorized directories | No |\n| `list_available_commands` | Show command categories | No |\n| `get_command_help` | Get command usage guidance | No |\n| `get_configuration` | View current settings | No |\n\n### Tool Examples\n\n#### Directory Management\n\n```python\n# Check available directories\ndirs = await list_directories(session_id=\"session123\")\nwhitelisted = dirs[\"whitelisted_directories\"]\napproved = dirs[\"session_approved_directories\"]\n\n# Request permission for a directory\nif \"/projects/my-data\" not in whitelisted and \"/projects/my-data\" not in approved:\n result = await approve_directory(\n directory=\"/projects/my-data\", \n session_id=\"session123\"\n )\n```\n\n#### Command Execution\n\n```python\n# Read commands (read permissions enforced)\nresult = await execute_read_command(\"ls -la ~/Documents\")\n\n# Any command type (may require command type approval)\nresult = await execute_command(\n command=\"mkdir -p ~/Projects/new-folder\", \n session_id=\"session123\"\n)\n```\n\n#### Get Configuration\n\n```python\n# Check current settings\nconfig = await get_configuration()\nwhitelist = config[\"directory_whitelisting\"][\"whitelisted_directories\"]\n```\n\n\n## Directory Security System\n\nThe server restricts command execution to specific directories, preventing access to sensitive files.\n\n### Directory Security Modes\n\nThe system supports three security modes:\n\n| Mode | Description | Best For | Configuration |\n|------|-------------|----------|--------------|\n| **Strict** | Only whitelisted directories allowed | Maximum security | `auto_approve_directories_in_desktop_mode: false` |\n| **Approval** | Non-whitelisted directories require explicit approval | Interactive use | Default behavior for standard clients |\n| **Auto-approve** | Auto-approves directories for Claude Desktop | Convenience | `auto_approve_directories_in_desktop_mode: true` |\n\n### Whitelisted Directory Configuration\n\n```json\n\"security\": {\n \"whitelisted_directories\": [\n \"/home\", // System directories\n \"/tmp\",\n \"~\", // User's home\n \"~/Documents\" // Common user directories\n ],\n \"auto_approve_directories_in_desktop_mode\": false // Set to true for convenience\n}\n```\n\n### Directory Approval Flow\n\n1. Command is requested in a directory\n2. System checks:\n - Is the directory in the global whitelist? → **Allow**\n - Has directory been approved in this session? → **Allow**\n - Neither? → **Request approval**\n3. After approval, directory remains approved for the entire session\n\n### Path Format Support\n\n- Absolute paths: `/home/user/documents`\n- Home directory: `~` (expands to user's home)\n- User subdirectories: `~/Downloads`\n\n### Claude Desktop Integration\n\nThe server maintains a persistent session for Claude Desktop, ensuring directory approvals persist between requests and preventing approval loops.\n\n## Command Customization\n\nThe system uses command categorization to control access:\n\n| Category | Description | Example Commands | Requires Approval |\n|----------|-------------|------------------|-------------------|\n| Read | Safe operations | ls, cat, find | No |\n| Write | Data modification | mkdir, rm, touch | Yes |\n| System | System operations | ps, ping, ifconfig | Yes |\n| Blocked | Dangerous commands | sudo, bash, eval | Always denied |\n\n### Customization Methods\n\n```json\n// In config.json\n{\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\", \"awk\", \"jq\"],\n \"write\": [\"mkdir\", \"touch\", \"rm\"],\n \"system\": [\"ping\", \"ifconfig\", \"kubectl\"],\n \"blocked\": [\"sudo\", \"bash\", \"eval\"]\n }\n}\n```\n\n**Environment Variable Method:**\n```bash\n# Add to existing lists, not replace (comma-separated)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq\"\nexport CMD_LINE_MCP_COMMANDS_BLOCKED=\"npm,pip\"\n```\n\nThe MCP server merges these additions with existing commands, letting you extend functionality without recreating complete command lists.\n\n### Command Chaining\n\nThe server supports three command chaining methods:\n\n| Method | Symbol | Example | Config Setting |\n|--------|--------|---------|---------------|\n| Pipes | `\\|` | `ls \\| grep txt` | `allow_command_separators: true` |\n| Sequence | `;` | `mkdir dir; cd dir` | `allow_command_separators: true` |\n| Background | `&` | `find . -name \"*.log\" &` | `allow_command_separators: true` |\n\nAll commands in a chain must be from the supported command list. Security checks apply to the entire chain.\n\n**Quick Configuration:**\n```json\n\"security\": {\n \"allow_command_separators\": true // Set to false to disable all chaining\n}\n```\n\nTo disable specific separators, add them to the `dangerous_patterns` list.\n\n## License\n\nMIT\n","English","tag","en-US",{"_29":41,"_33":45,"_38":46},{"_31":42,"_33":43,"_35":44},"Permite que los asistentes de IA ejecuten de forma segura comandos de terminal comunes de Unix/macOS a través de una interfaz controlada con funciones de seguridad integrales.","Servidor MCP de línea de comandos","# Servidor MCP de línea de comandos\n\n[![Versión de PyPI](https://badge.fury.io/py/cmd-line-mcp.svg)](https://badge.fury.io/py/cmd-line-mcp) [![Versiones de Python](https://img.shields.io/pypi/pyversions/cmd-line-mcp.svg)](https://pypi.org/project/cmd-line-mcp/) [![Licencia: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\nUn servidor de Protocolo de Control de Modelos (MCP) seguro que permite a los asistentes de IA ejecutar comandos de terminal con acceso a directorios y permisos de comando controlados.\n\n## Descripción general\n\nEl MCP de línea de comandos proporciona una capa de seguridad entre los asistentes de IA y su terminal. Implementa un modelo de seguridad dual:\n\n1. **Permisos de comando** : Los comandos se clasifican como lectura (seguro), escritura (cambia datos) o sistema (afecta el estado del sistema), con diferentes requisitos de aprobación.\n2. **Permisos de directorio** : los comandos solo pueden acceder a directorios incluidos explícitamente en la lista blanca o directorios aprobados durante una sesión\n\nLos asistentes de IA interactúan con este servidor mediante herramientas MCP estandarizadas, lo que permite la ejecución segura de comandos de terminal y evita el acceso a archivos confidenciales o a operaciones peligrosas. Puede configurar el nivel de seguridad, desde altamente restrictivo hasta más permisivo, según sus necesidades.\n\n## Características principales\n\n| Seguridad | Usabilidad | Integración |\n| ----------------------------- | ------------------------------------------------------ | ---------------------------------- |\n| Lista blanca de directorios | Categorización de comandos (lectura/escritura/sistema) | Compatibilidad con Claude Desktop |\n| Filtrado de comandos | Permisos de sesión persistentes | Protocolo MCP estándar |\n| Coincidencia de patrones | Encadenamiento de comandos (tuberías, etc.) | Opciones de aprobación automática |\n| Bloqueo de comandos peligroso | Flujo de trabajo de aprobación intuitivo | Múltiples métodos de configuración |\n\n## Comandos compatibles (listos para usar)\n\n### Comandos de lectura\n\n* `ls` , `pwd` , `cat` , `less` , `head` , `tail` , `grep` , `find` , `which` , `du` , `df` , `file` , `sort` , etc.\n\n### Comandos de escritura\n\n* `cp` , `mv` , `rm` , `mkdir` , `rmdir` , `touch` , `chmod` , `chown` , etc.\n\n### Comandos del sistema\n\n* `ps` , `top` , `htop` , `who` , `netstat` , `ifconfig` , `ping` , etc.\n\n## Arquitectura de seguridad\n\nEl sistema implementa un enfoque de seguridad de múltiples capas:\n\n```\n┌───────────────────────────────────────────────────────────────┐\n│ COMMAND-LINE MCP SERVER │\n├──────────────────┬────────────────────────┬───────────────────┤\n│ COMMAND SECURITY │ DIRECTORY SECURITY │ SESSION SECURITY │\n├──────────────────┼────────────────────────┼───────────────────┤\n│ ✓ Read commands │ ✓ Directory whitelist │ ✓ Session IDs │\n│ ✓ Write commands │ ✓ Runtime approvals │ ✓ Persistent │\n│ ✓ System commands│ ✓ Path validation │ permissions │\n│ ✓ Blocked list │ ✓ Home dir expansion │ ✓ Auto timeouts │\n│ ✓ Pattern filters│ ✓ Subdirectory check │ ✓ Desktop mode │\n└──────────────────┴────────────────────────┴───────────────────┘\n```\n\nTodas las funciones de seguridad se pueden configurar de restrictivas a permisivas según su modelo de amenaza y sus requisitos de conveniencia.\n\n## Inicio rápido\n\n```bash\n# Install\ngit clone https://github.com/yourusername/cmd-line-mcp.git\ncd cmd-line-mcp\npython -m venv venv\nsource venv/bin/activate\npip install -e .\ncp config.json.example config.json\n\n# Run\ncmd-line-mcp # With default config\ncmd-line-mcp --config config.json # With specific config\n```\n\n### Opciones de configuración\n\nEl servidor admite cuatro métodos de configuración en orden de precedencia:\n\n1. **Configuración predeterminada incorporada** (default\\_config.json)\n\n2. **Archivo de configuración JSON** (recomendado para personalización)\n\n ```bash\n cmd-line-mcp --config config.json\n ```\n\n3. **Variables de entorno** (para modificaciones específicas)\n\n ```bash\n export CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"~,/tmp\"\n ```\n\n4. **Archivo .env** (para configuraciones específicas del entorno)\n\n ```bash\n cmd-line-mcp --config config.json --env .env\n ```\n\nLa configuración predeterminada se almacena en `default_config.json` y se incluye en el paquete. Puedes copiar este archivo para crear tu propia configuración personalizada.\n\n#### Configuración básica\n\n```json\n{\n \"security\": {\n \"whitelisted_directories\": [\"/home\", \"/tmp\", \"~\"],\n \"auto_approve_directories_in_desktop_mode\": false, \n \"require_session_id\": false,\n \"allow_command_separators\": true\n },\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\"], \n \"write\": [\"touch\", \"mkdir\", \"rm\"],\n \"system\": [\"ps\", \"ping\"]\n }\n}\n```\n\n#### Formato de variable de entorno\n\nLas variables de entorno utilizan un patrón de nombres predecible:\n\n```\nCMD_LINE_MCP_

_\n```\n\nEjemplos:\n\n```bash\n# Security settings\nexport CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"/projects,/var/data\"\nexport CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE=true\n\n# Command additions (these merge with defaults)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq,wc\"\n```\n\n### Integración de escritorio de Claude\n\n#### Configuración\n\n1. Instalar [Claude para escritorio](https://claude.ai/download)\n2. Configurar en `~/Library/Application Support/Claude/claude_desktop_config.json` :\n\n```json\n{\n \"mcpServers\": {\n \"cmd-line\": {\n \"command\": \"/path/to/venv/bin/cmd-line-mcp\",\n \"args\": [\"--config\", \"/path/to/config.json\"],\n \"env\": {\n \"CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID\": \"false\",\n \"CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE\": \"true\"\n }\n }\n }\n}\n```\n\n#### Configuración recomendada de escritorio de Claude\n\nPara una mejor experiencia, configure:\n\n* `require_session_id: false` - Esencial para evitar bucles de aprobación\n* `auto_approve_directories_in_desktop_mode: true` - Opcional para un acceso conveniente\n* Incluya directorios comunes en su lista blanca\n\nDespués de la configuración, reinicie Claude for Desktop.\n\n## Herramientas de asistencia de IA\n\nEl servidor proporciona estas herramientas MCP para asistentes de IA:\n\n| Herramienta | Objetivo | Necesita aprobación |\n| ------------------------- | -------------------------------------------- | -------------------------------------- |\n| `execute_command` | Ejecutar cualquier tipo de comando | Sí, para comandos de escritura/sistema |\n| `execute_read_command` | Ejecutar comandos de solo lectura | Sólo aprobación del directorio |\n| `approve_directory` | Otorgar acceso a un directorio | N/A - es una herramienta de aprobación |\n| `approve_command_type` | Otorgar permiso para la categoría de comando | N/A - es una herramienta de aprobación |\n| `list_directories` | Mostrar directorios autorizados | No |\n| `list_available_commands` | Mostrar categorías de comandos | No |\n| `get_command_help` | Obtenga orientación sobre el uso de comandos | No |\n| `get_configuration` | Ver la configuración actual | No |\n\n### Ejemplos de herramientas\n\n#### Gestión de directorios\n\n```python\n# Check available directories\ndirs = await list_directories(session_id=\"session123\")\nwhitelisted = dirs[\"whitelisted_directories\"]\napproved = dirs[\"session_approved_directories\"]\n\n# Request permission for a directory\nif \"/projects/my-data\" not in whitelisted and \"/projects/my-data\" not in approved:\n result = await approve_directory(\n directory=\"/projects/my-data\", \n session_id=\"session123\"\n )\n```\n\n#### Ejecución de comandos\n\n```python\n# Read commands (read permissions enforced)\nresult = await execute_read_command(\"ls -la ~/Documents\")\n\n# Any command type (may require command type approval)\nresult = await execute_command(\n command=\"mkdir -p ~/Projects/new-folder\", \n session_id=\"session123\"\n)\n```\n\n#### Obtener configuración\n\n```python\n# Check current settings\nconfig = await get_configuration()\nwhitelist = config[\"directory_whitelisting\"][\"whitelisted_directories\"]\n```\n\n## Sistema de seguridad de directorios\n\nEl servidor restringe la ejecución de comandos a directorios específicos, evitando el acceso a archivos confidenciales.\n\n### Modos de seguridad de directorio\n\nEl sistema admite tres modos de seguridad:\n\n| Modo | Descripción | Mejor para | Configuración |\n| ------------------------- | ------------------------------------------------------------------------------ | ---------------- | ---------------------------------------------------- |\n| **Estricto** | Solo se permiten directorios incluidos en la lista blanca | Máxima seguridad | `auto_approve_directories_in_desktop_mode: false` |\n| **Aprobación** | Los directorios no incluidos en la lista blanca requieren aprobación explícita | Uso interactivo | Comportamiento predeterminado para clientes estándar |\n| **Aprobación automática** | Aprueba automáticamente directorios para Claude Desktop | Conveniencia | `auto_approve_directories_in_desktop_mode: true` |\n\n### Configuración de directorio en lista blanca\n\n```json\n\"security\": {\n \"whitelisted_directories\": [\n \"/home\", // System directories\n \"/tmp\",\n \"~\", // User's home\n \"~/Documents\" // Common user directories\n ],\n \"auto_approve_directories_in_desktop_mode\": false // Set to true for convenience\n}\n```\n\n### Flujo de aprobación del directorio\n\n1. Se solicita el comando en un directorio\n2. Comprobaciones del sistema:\n * ¿Está el directorio en la lista blanca global? → **Permitir**\n * ¿Se ha aprobado el directorio en esta sesión? → **Permitir**\n * ¿Ninguno? → **Solicitar aprobación**\n3. Después de la aprobación, el directorio permanece aprobado durante toda la sesión.\n\n### Compatibilidad con formatos de ruta\n\n* Rutas absolutas: `/home/user/documents`\n* Directorio de inicio: `~` (se expande al inicio del usuario)\n* Subdirectorios de usuario: `~/Downloads`\n\n### Integración de escritorio de Claude\n\nEl servidor mantiene una sesión persistente para Claude Desktop, lo que garantiza que las aprobaciones de directorio persistan entre solicitudes y evita bucles de aprobación.\n\n## Personalización de comandos\n\nEl sistema utiliza la categorización de comandos para controlar el acceso:\n\n| Categoría | Descripción | Comandos de ejemplo | Requiere aprobación |\n| --------- | ----------------------- | ------------------- | ------------------- |\n| Leer | Operaciones seguras | ls, gato, encontrar | No |\n| Escribir | Modificación de datos | mkdir, rm, tocar | Sí |\n| Sistema | Operaciones del sistema | ps, ping, ifconfig | Sí |\n| Obstruido | Órdenes peligrosas | sudo, bash, eval | Siempre negado |\n\n### Métodos de personalización\n\n```json\n// In config.json\n{\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\", \"awk\", \"jq\"],\n \"write\": [\"mkdir\", \"touch\", \"rm\"],\n \"system\": [\"ping\", \"ifconfig\", \"kubectl\"],\n \"blocked\": [\"sudo\", \"bash\", \"eval\"]\n }\n}\n```\n\n**Método de variable de entorno:**\n\n```bash\n# Add to existing lists, not replace (comma-separated)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq\"\nexport CMD_LINE_MCP_COMMANDS_BLOCKED=\"npm,pip\"\n```\n\nEl servidor MCP fusiona estas adiciones con comandos existentes, lo que le permite ampliar la funcionalidad sin tener que recrear listas de comandos completas.\n\n### Encadenamiento de comandos\n\nEl servidor admite tres métodos de encadenamiento de comandos:\n\n| Método | Símbolo | Ejemplo | Configuración | | |\n| --------- | ------- | ------------------------ | -------------------------------- | ---------- | -------------------------------- |\n| Tubería | \\` | \\` | \\`ls | grep txt\\` | `allow_command_separators: true` |\n| Secuencia | `;` | `mkdir dir; cd dir` | `allow_command_separators: true` | | |\n| Fondo | `&` | `find . -name \"*.log\" &` | `allow_command_separators: true` | | |\n\nTodos los comandos de una cadena deben pertenecer a la lista de comandos compatibles. Las comprobaciones de seguridad se aplican a toda la cadena.\n\n**Configuración rápida:**\n\n```json\n\"security\": {\n \"allow_command_separators\": true // Set to false to disable all chaining\n}\n```\n\nPara deshabilitar separadores específicos, agréguelos a la lista `dangerous_patterns` .\n\n## Licencia\n\nInstituto Tecnológico de Massachusetts (MIT)","Spanish","es-ES",{"_29":48,"_33":52,"_38":53},{"_31":49,"_33":50,"_35":51},"AI アシスタントが、包括的なセキュリティ機能を備えた制御されたインターフェースを通じて、一般的な Unix/macOS ターミナルコマンドを安全に実行できるようにします。","コマンドライン MCP サーバー","# コマンドライン MCP サーバー\n\n[![PyPIバージョン](https://badge.fury.io/py/cmd-line-mcp.svg)](https://badge.fury.io/py/cmd-line-mcp) [![Pythonのバージョン](https://img.shields.io/pypi/pyversions/cmd-line-mcp.svg)](https://pypi.org/project/cmd-line-mcp/) [![ライセンス: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\nAI アシスタントが制御されたディレクトリアクセスとコマンド権限を使用してターミナルコマンドを実行できるようにする、安全なモデル制御プロトコル (MCP) サーバー。\n\n## 概要\n\nコマンドラインMCPは、AIアシスタントと端末の間にセキュリティレイヤーを提供します。二重のセキュリティモデルを実装しています。\n\n1. **コマンド権限**: コマンドは、読み取り（安全）、書き込み（データの変更）、システム（システム状態に影響）に分類され、承認要件が異なります。\n2. **ディレクトリ権限**: コマンドは、明示的にホワイトリストに登録されたディレクトリまたはセッション中に承認されたディレクトリにのみアクセスできます。\n\nAIアシスタントは標準化されたMCPツールを使用してこのサーバーと連携し、機密ファイルへのアクセスや危険な操作を防止しながら、安全なターミナルコマンド実行を可能にします。セキュリティレベルは、ニーズに応じて、非常に制限の厳しいものからより緩やかなものまで設定できます。\n\n## 主な特徴\n\n| 安全 | ユーザビリティ | 統合 |\n| -------------- | ----------------------- | ------------- |\n| ディレクトリのホワイトリスト | コマンドの分類（読み取り/書き込み/システム） | クロードデスクトップ互換性 |\n| コマンドフィルタリング | 永続的なセッション権限 | 標準MCPプロトコル |\n| パターンマッチング | コマンド連鎖（パイプなど） | 自動承認オプション |\n| 危険なコマンドブロッキング | 直感的な承認ワークフロー | 複数の設定方法 |\n\n## サポートされているコマンド（すぐに使用可能）\n\n### 読み取りコマンド\n\n* `ls` 、 `pwd` 、 `cat` 、 `less` 、 `head` 、 `tail` 、 `grep` 、 `find` 、 `which` 、 `du` 、 `df` 、 `file` 、 `sort`など。\n\n### 書き込みコマンド\n\n* `cp` 、 `mv` 、 `rm` 、 `mkdir` 、 `rmdir` 、 `touch` 、 `chmod` 、 `chown`など。\n\n### システムコマンド\n\n* `ps` 、 `top` 、 `htop` 、 `who` 、 `netstat` 、 `ifconfig` 、 `ping`など。\n\n## セキュリティアーキテクチャ\n\nこのシステムは多層セキュリティアプローチを実装しています。\n\n```\n┌───────────────────────────────────────────────────────────────┐\n│ COMMAND-LINE MCP SERVER │\n├──────────────────┬────────────────────────┬───────────────────┤\n│ COMMAND SECURITY │ DIRECTORY SECURITY │ SESSION SECURITY │\n├──────────────────┼────────────────────────┼───────────────────┤\n│ ✓ Read commands │ ✓ Directory whitelist │ ✓ Session IDs │\n│ ✓ Write commands │ ✓ Runtime approvals │ ✓ Persistent │\n│ ✓ System commands│ ✓ Path validation │ permissions │\n│ ✓ Blocked list │ ✓ Home dir expansion │ ✓ Auto timeouts │\n│ ✓ Pattern filters│ ✓ Subdirectory check │ ✓ Desktop mode │\n└──────────────────┴────────────────────────┴───────────────────┘\n```\n\nすべてのセキュリティ機能は、脅威モデルと利便性の要件に基づいて、制限付きから許可付きまで構成できます。\n\n## クイックスタート\n\n```bash\n# Install\ngit clone https://github.com/yourusername/cmd-line-mcp.git\ncd cmd-line-mcp\npython -m venv venv\nsource venv/bin/activate\npip install -e .\ncp config.json.example config.json\n\n# Run\ncmd-line-mcp # With default config\ncmd-line-mcp --config config.json # With specific config\n```\n\n### 設定オプション\n\nサーバーは、優先順位に従って 4 つの構成方法をサポートしています。\n\n1. **組み込みのデフォルト構成**(default\\_config.json)\n\n2. **JSON 構成ファイル**(カスタマイズに推奨)\n\n ```bash\n cmd-line-mcp --config config.json\n ```\n\n3. **環境変数**（特定のオーバーライド用）\n\n ```bash\n export CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"~,/tmp\"\n ```\n\n4. **.env ファイル**（環境固有の設定用）\n\n ```bash\n cmd-line-mcp --config config.json --env .env\n ```\n\nデフォルト設定は`default_config.json`に保存されており、パッケージに含まれています。このファイルをコピーして、独自のカスタム設定を作成できます。\n\n#### コア構成設定\n\n```json\n{\n \"security\": {\n \"whitelisted_directories\": [\"/home\", \"/tmp\", \"~\"],\n \"auto_approve_directories_in_desktop_mode\": false, \n \"require_session_id\": false,\n \"allow_command_separators\": true\n },\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\"], \n \"write\": [\"touch\", \"mkdir\", \"rm\"],\n \"system\": [\"ps\", \"ping\"]\n }\n}\n```\n\n#### 環境変数の形式\n\n環境変数では予測可能な命名パターンが使用されます。\n\n```\nCMD_LINE_MCP_

_\n```\n\n例:\n\n```bash\n# Security settings\nexport CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"/projects,/var/data\"\nexport CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE=true\n\n# Command additions (these merge with defaults)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq,wc\"\n```\n\n### クロードデスクトップ統合\n\n#### 設定\n\n1. [デスクトップ版Claude](https://claude.ai/download)をインストールする\n2. `~/Library/Application Support/Claude/claude_desktop_config.json`で設定します:\n\n```json\n{\n \"mcpServers\": {\n \"cmd-line\": {\n \"command\": \"/path/to/venv/bin/cmd-line-mcp\",\n \"args\": [\"--config\", \"/path/to/config.json\"],\n \"env\": {\n \"CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID\": \"false\",\n \"CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE\": \"true\"\n }\n }\n }\n}\n```\n\n#### 推奨されるクロードデスクトップ設定\n\n最適なエクスペリエンスを得るには、以下を設定してください。\n\n* `require_session_id: false` - 承認ループを防ぐために必須\n* `auto_approve_directories_in_desktop_mode: true` - アクセスを容易にするためのオプション\n* 一般的なディレクトリをホワイトリストに含める\n\n設定後、Claude for Desktop を再起動します。\n\n## AIアシスタントツール\n\nサーバーは、AI アシスタント用に次の MCP ツールを提供します。\n\n| 道具 | 目的 | 承認が必要 |\n| ------------------------- | ----------------- | ------------------- |\n| `execute_command` | 任意のコマンドタイプを実行する | はい、書き込み/システムコマンドの場合 |\n| `execute_read_command` | 読み取り専用コマンドを実行する | ディレクトリ承認のみ |\n| `approve_directory` | ディレクトリへのアクセスを許可する | N/A - 承認ツールです |\n| `approve_command_type` | コマンドカテゴリの権限を付与する | N/A - 承認ツールです |\n| `list_directories` | 承認されたディレクトリを表示 | いいえ |\n| `list_available_commands` | コマンドカテゴリを表示 | いいえ |\n| `get_command_help` | コマンドの使用ガイドを取得する | いいえ |\n| `get_configuration` | 現在の設定を表示 | いいえ |\n\n### ツールの例\n\n#### ディレクトリ管理\n\n```python\n# Check available directories\ndirs = await list_directories(session_id=\"session123\")\nwhitelisted = dirs[\"whitelisted_directories\"]\napproved = dirs[\"session_approved_directories\"]\n\n# Request permission for a directory\nif \"/projects/my-data\" not in whitelisted and \"/projects/my-data\" not in approved:\n result = await approve_directory(\n directory=\"/projects/my-data\", \n session_id=\"session123\"\n )\n```\n\n#### コマンド実行\n\n```python\n# Read commands (read permissions enforced)\nresult = await execute_read_command(\"ls -la ~/Documents\")\n\n# Any command type (may require command type approval)\nresult = await execute_command(\n command=\"mkdir -p ~/Projects/new-folder\", \n session_id=\"session123\"\n)\n```\n\n#### 構成を取得\n\n```python\n# Check current settings\nconfig = await get_configuration()\nwhitelist = config[\"directory_whitelisting\"][\"whitelisted_directories\"]\n```\n\n## ディレクトリセキュリティシステム\n\nサーバーはコマンドの実行を特定のディレクトリに制限し、機密ファイルへのアクセスを防止します。\n\n### ディレクトリセキュリティモード\n\nシステムは次の 3 つのセキュリティモードをサポートしています。\n\n| モード | 説明 | 最適な用途 | 構成 |\n| -------- | ---------------------------------- | ----------- | ------------------------------------------------- |\n| **厳しい** | ホワイトリストに登録されたディレクトリのみ許可 | 最大限のセキュリティ | `auto_approve_directories_in_desktop_mode: false` |\n| **承認** | ホワイトリストに登録されていないディレクトリは明示的な承認が必要です | インタラクティブな使用 | 標準クライアントのデフォルトの動作 |\n| **自動承認** | Claude Desktop のディレクトリを自動承認します | 利便性 | `auto_approve_directories_in_desktop_mode: true` |\n\n### ホワイトリストディレクトリ構成\n\n```json\n\"security\": {\n \"whitelisted_directories\": [\n \"/home\", // System directories\n \"/tmp\",\n \"~\", // User's home\n \"~/Documents\" // Common user directories\n ],\n \"auto_approve_directories_in_desktop_mode\": false // Set to true for convenience\n}\n```\n\n### ディレクトリ承認フロー\n\n1. コマンドはディレクトリ内で要求されます\n2. システムチェック:\n * ディレクトリはグローバルホワイトリストに登録されていますか？→**許可**\n * このセッションでディレクトリは承認されましたか? →**許可**\n * どちらでもない？→**承認をリクエスト**\n3. 承認後、ディレクトリはセッション全体にわたって承認済みのままになります\n\n### パス形式のサポート\n\n* 絶対パス: `/home/user/documents`\n* ホームディレクトリ: `~` (ユーザーのホームに展開)\n* ユーザーのサブディレクトリ: `~/Downloads`\n\n### クロードデスクトップ統合\n\nサーバーは Claude Desktop の永続的なセッションを維持し、ディレクトリ承認がリクエスト間で維持され、承認ループが防止されるようにします。\n\n## コマンドのカスタマイズ\n\nシステムはコマンド分類を使用してアクセスを制御します。\n\n| カテゴリ | 説明 | コマンド例 | 承認が必要 |\n| --------- | ------- | ---------------- | ------- |\n| 読む | 安全な操作 | ls、cat、find | いいえ |\n| 書く | データの変更 | mkdir、rm、touch | はい |\n| システム | システム操作 | ps、ping、ifconfig | はい |\n| ブロックされました | 危険なコマンド | sudo、bash、eval | 常に拒否される |\n\n### カスタマイズ方法\n\n```json\n// In config.json\n{\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\", \"awk\", \"jq\"],\n \"write\": [\"mkdir\", \"touch\", \"rm\"],\n \"system\": [\"ping\", \"ifconfig\", \"kubectl\"],\n \"blocked\": [\"sudo\", \"bash\", \"eval\"]\n }\n}\n```\n\n**環境変数方式:**\n\n```bash\n# Add to existing lists, not replace (comma-separated)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq\"\nexport CMD_LINE_MCP_COMMANDS_BLOCKED=\"npm,pip\"\n```\n\nMCP サーバーはこれらの追加機能を既存のコマンドと結合し、完全なコマンドリストを再作成せずに機能を拡張できるようにします。\n\n### コマンドチェーン\n\nサーバーは、次の 3 つのコマンド連鎖方式をサポートしています。\n\n| 方法 | シンボル | 例 | 設定 | | |\n| --- | ---- | ------------------------ | -------------------------------- | ---------- | -------------------------------- |\n| パイプ | \\` | \\` | \\`ls | grep txt\\` | `allow_command_separators: true` |\n| 順序 | `;` | `mkdir dir; cd dir` | `allow_command_separators: true` | | |\n| 背景 | `&` | `find . -name \"*.log\" &` | `allow_command_separators: true` | | |\n\nチェーン内のすべてのコマンドは、サポートされているコマンドリストから選択する必要があります。セキュリティチェックはチェーン全体に適用されます。\n\n**クイック設定:**\n\n```json\n\"security\": {\n \"allow_command_separators\": true // Set to false to disable all chaining\n}\n```\n\n特定のセパレーターを無効にするには、それらを`dangerous_patterns`リストに追加します。\n\n## ライセンス\n\nマサチューセッツ工科大学","Japanese","ja-JP",{"_29":55,"_33":59,"_38":60},{"_31":56,"_33":57,"_35":58},"AI 보조원이 포괄적인 보안 기능을 갖춘 제어된 인터페이스를 통해 일반적인 Unix/macOS 터미널 명령을 안전하게 실행할 수 있도록 합니다.","명령줄 MCP 서버","# 명령줄 MCP 서버\n\n[![PyPI 버전](https://badge.fury.io/py/cmd-line-mcp.svg)](https://badge.fury.io/py/cmd-line-mcp) [![파이썬 버전](https://img.shields.io/pypi/pyversions/cmd-line-mcp.svg)](https://pypi.org/project/cmd-line-mcp/) [![라이센스: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\nAI 어시스턴트가 제어된 디렉토리 액세스 및 명령 권한을 통해 터미널 명령을 실행할 수 있도록 하는 안전한 MCP(Model Control Protocol) 서버입니다.\n\n## 개요\n\n명령줄 MCP는 AI 어시스턴트와 단말기 사이에 보안 계층을 제공합니다. 다음과 같은 이중 보안 모델을 구현합니다.\n\n1. **명령 권한** : 명령은 읽기(안전), 쓰기(데이터 변경), 시스템(시스템 상태에 영향을 미침)으로 분류되며 승인 요구 사항이 다릅니다.\n2. **디렉토리 권한** : 명령은 세션 중에 승인된 디렉토리 또는 명시적으로 허용된 디렉토리에만 액세스할 수 있습니다.\n\nAI 어시스턴트는 표준화된 MCP 도구를 사용하여 이 서버와 상호 작용하여 민감한 파일이나 위험한 작업에 대한 접근을 차단하는 동시에 안전한 터미널 명령 실행을 지원합니다. 필요에 따라 매우 제한적인 보안 수준부터 보다 관대한 보안 수준까지 보안 수준을 설정할 수 있습니다.\n\n## 주요 특징\n\n| 보안 | 사용성 | 완성 |\n| ---------- | ----------------- | ------------------ |\n| 디렉토리 허용 목록 | 명령어 분류(읽기/쓰기/시스템) | Claude Desktop 호환성 |\n| 명령 필터링 | 지속적인 세션 권한 | 표준 MCP 프로토콜 |\n| 패턴 매칭 | 명령 체이닝(파이프 등) | 자동 승인 옵션 |\n| 위험한 명령 차단 | 직관적인 승인 워크플로 | 여러 구성 방법 |\n\n## 지원되는 명령(기본 제공)\n\n### 명령 읽기\n\n* `ls` , `pwd` , `cat` , `less` , `head` , `tail` , `grep` , `find` , `which` , `du` , `df` , `file` , `sort` 등.\n\n### 쓰기 명령\n\n* `cp` , `mv` , `rm` , `mkdir` , `rmdir` , `touch` , `chmod` , `chown` 등.\n\n### 시스템 명령\n\n* `ps` , `top` , `htop` , `who` , `netstat` , `ifconfig` , `ping` 등\n\n## 보안 아키텍처\n\n시스템은 다층 보안 접근 방식을 구현합니다.\n\n지엑스피1\n\n모든 보안 기능은 위협 모델과 편의성 요구 사항에 따라 제한적에서 허용적까지 구성할 수 있습니다.\n\n## 빠른 시작\n\n```bash\n# Install\ngit clone https://github.com/yourusername/cmd-line-mcp.git\ncd cmd-line-mcp\npython -m venv venv\nsource venv/bin/activate\npip install -e .\ncp config.json.example config.json\n\n# Run\ncmd-line-mcp # With default config\ncmd-line-mcp --config config.json # With specific config\n```\n\n### 구성 옵션\n\n서버는 우선순위에 따라 4가지 구성 방법을 지원합니다.\n\n1. **기본 구성** (default\\_config.json)\n\n2. **JSON 구성 파일** (사용자 정의에 권장됨)\n\n ```bash\n cmd-line-mcp --config config.json\n ```\n\n3. **환경 변수** (특정 재정의용)\n\n ```bash\n export CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"~,/tmp\"\n ```\n\n4. **.env 파일** (환경별 설정용)\n\n ```bash\n cmd-line-mcp --config config.json --env .env\n ```\n\n기본 구성은 `default_config.json` 에 저장되어 있으며 패키지에 포함되어 있습니다. 이 파일을 복사하여 사용자 지정 구성을 만들 수 있습니다.\n\n#### 핵심 구성 설정\n\n```json\n{\n \"security\": {\n \"whitelisted_directories\": [\"/home\", \"/tmp\", \"~\"],\n \"auto_approve_directories_in_desktop_mode\": false, \n \"require_session_id\": false,\n \"allow_command_separators\": true\n },\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\"], \n \"write\": [\"touch\", \"mkdir\", \"rm\"],\n \"system\": [\"ps\", \"ping\"]\n }\n}\n```\n\n#### 환경 변수 형식\n\n환경 변수는 예측 가능한 명명 패턴을 사용합니다.\n\n```\nCMD_LINE_MCP_

_\n```\n\n예:\n\n```bash\n# Security settings\nexport CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"/projects,/var/data\"\nexport CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE=true\n\n# Command additions (these merge with defaults)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq,wc\"\n```\n\n### Claude 데스크톱 통합\n\n#### 설정\n\n1. [데스크톱용 Claude](https://claude.ai/download) 설치\n2. `~/Library/Application Support/Claude/claude_desktop_config.json` 에서 구성합니다.\n\n```json\n{\n \"mcpServers\": {\n \"cmd-line\": {\n \"command\": \"/path/to/venv/bin/cmd-line-mcp\",\n \"args\": [\"--config\", \"/path/to/config.json\"],\n \"env\": {\n \"CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID\": \"false\",\n \"CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE\": \"true\"\n }\n }\n }\n}\n```\n\n#### 권장되는 Claude 데스크톱 설정\n\n최상의 환경을 위해 다음을 구성하세요.\n\n* `require_session_id: false` - 승인 루프를 방지하는 데 필수\n* `auto_approve_directories_in_desktop_mode: true` - 편리한 액세스를 위한 선택 사항\n* 허용 목록에 일반 디렉토리를 포함하세요\n\n구성 후 Claude for Desktop을 다시 시작합니다.\n\n## AI 어시스턴트 도구\n\n서버는 AI 어시스턴트를 위해 다음과 같은 MCP 도구를 제공합니다.\n\n| 도구 | 목적 | 승인이 필요합니다 |\n| ------------------------- | ------------------ | ---------------- |\n| `execute_command` | 모든 명령 유형을 실행하세요 | 예, 쓰기/시스템 명령의 경우 |\n| `execute_read_command` | 읽기 전용 명령 실행 | 디렉토리 승인만 가능 |\n| `approve_directory` | 디렉토리에 대한 액세스 권한 부여 | N/A - 승인 도구입니다. |\n| `approve_command_type` | 명령 범주에 대한 권한 부여 | N/A - 승인 도구입니다. |\n| `list_directories` | 승인된 디렉토리 표시 | 아니요 |\n| `list_available_commands` | 명령 범주 표시 | 아니요 |\n| `get_command_help` | 명령어 사용 지침 받기 | 아니요 |\n| `get_configuration` | 현재 설정 보기 | 아니요 |\n\n### 도구 예제\n\n#### 디렉토리 관리\n\n```python\n# Check available directories\ndirs = await list_directories(session_id=\"session123\")\nwhitelisted = dirs[\"whitelisted_directories\"]\napproved = dirs[\"session_approved_directories\"]\n\n# Request permission for a directory\nif \"/projects/my-data\" not in whitelisted and \"/projects/my-data\" not in approved:\n result = await approve_directory(\n directory=\"/projects/my-data\", \n session_id=\"session123\"\n )\n```\n\n#### 명령 실행\n\n```python\n# Read commands (read permissions enforced)\nresult = await execute_read_command(\"ls -la ~/Documents\")\n\n# Any command type (may require command type approval)\nresult = await execute_command(\n command=\"mkdir -p ~/Projects/new-folder\", \n session_id=\"session123\"\n)\n```\n\n#### 구성 가져오기\n\n```python\n# Check current settings\nconfig = await get_configuration()\nwhitelist = config[\"directory_whitelisting\"][\"whitelisted_directories\"]\n```\n\n## 디렉토리 보안 시스템\n\n서버는 명령 실행을 특정 디렉토리로 제한하여 중요한 파일에 대한 액세스를 차단합니다.\n\n### 디렉토리 보안 모드\n\n이 시스템은 세 가지 보안 모드를 지원합니다.\n\n| 방법 | 설명 | 가장 적합한 | 구성 |\n| --------- | -------------------------------- | ------ | ------------------------------------------------- |\n| **엄격한** | 허용 목록에 있는 디렉토리만 허용됩니다. | 최대 보안 | `auto_approve_directories_in_desktop_mode: false` |\n| **승인** | 허용 목록에 없는 디렉토리에는 명시적인 승인이 필요합니다. | 대화형 사용 | 표준 클라이언트의 기본 동작 |\n| **자동 승인** | Claude Desktop에 대한 디렉토리 자동 승인 | 편의 | `auto_approve_directories_in_desktop_mode: true` |\n\n### 허용 목록에 있는 디렉토리 구성\n\n```json\n\"security\": {\n \"whitelisted_directories\": [\n \"/home\", // System directories\n \"/tmp\",\n \"~\", // User's home\n \"~/Documents\" // Common user directories\n ],\n \"auto_approve_directories_in_desktop_mode\": false // Set to true for convenience\n}\n```\n\n### 디렉토리 승인 흐름\n\n1. 디렉토리에서 명령이 요청되었습니다.\n2. 시스템 검사:\n * 해당 디렉토리가 글로벌 허용 목록에 있나요? → **허용**\n * 이번 세션에서 디렉토리가 승인되었습니까? → **허용**\n * 둘 다 아닌가요? → **승인 요청**\n3. 승인 후 디렉토리는 세션 전체에 걸쳐 승인된 상태로 유지됩니다.\n\n### 경로 형식 지원\n\n* 절대 경로: `/home/user/documents`\n* 홈 디렉토리: `~` (사용자 홈으로 확장)\n* 사용자 하위 디렉토리: `~/Downloads`\n\n### Claude 데스크톱 통합\n\n서버는 Claude Desktop에 대한 지속적인 세션을 유지하여 요청 간에 디렉토리 승인이 유지되도록 하고 승인 루프를 방지합니다.\n\n## 명령 사용자 정의\n\n시스템은 명령 분류를 사용하여 액세스를 제어합니다.\n\n| 범주 | 설명 | 예제 명령 | 승인이 필요합니다 |\n| -- | ------ | ------------------ | --------- |\n| 읽다 | 안전한 작업 | ls, 고양이, 찾기 | 아니요 |\n| 쓰다 | 데이터 수정 | mkdir, rm, touch | 예 |\n| 체계 | 시스템 운영 | ps, ping, ifconfig | 예 |\n| 막힌 | 위험한 명령 | sudo, bash, eval | 항상 거부당함 |\n\n### 사용자 정의 방법\n\n```json\n// In config.json\n{\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\", \"awk\", \"jq\"],\n \"write\": [\"mkdir\", \"touch\", \"rm\"],\n \"system\": [\"ping\", \"ifconfig\", \"kubectl\"],\n \"blocked\": [\"sudo\", \"bash\", \"eval\"]\n }\n}\n```\n\n**환경 변수 방법:**\n\n```bash\n# Add to existing lists, not replace (comma-separated)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq\"\nexport CMD_LINE_MCP_COMMANDS_BLOCKED=\"npm,pip\"\n```\n\nMCP 서버는 이러한 추가 기능을 기존 명령과 병합하여 전체 명령 목록을 다시 만들지 않고도 기능을 확장할 수 있도록 해줍니다.\n\n### 명령 체이닝\n\n서버는 세 가지 명령 체이닝 방법을 지원합니다.\n\n| 방법 | 상징 | 예 | 구성 설정 | | |\n| --- | --- | ------------------------ | -------------------------------- | ---------- | -------------------------------- |\n| 파이프 | \\` | \\` | \\`ls | grep txt\\` | `allow_command_separators: true` |\n| 순서 | `;` | `mkdir dir; cd dir` | `allow_command_separators: true` | | |\n| 배경 | `&` | `find . -name \"*.log\" &` | `allow_command_separators: true` | | |\n\n체인에 있는 모든 명령은 지원되는 명령 목록에 속해야 합니다. 보안 검사는 체인 전체에 적용됩니다.\n\n**빠른 구성:**\n\n```json\n\"security\": {\n \"allow_command_separators\": true // Set to false to disable all chaining\n}\n```\n\n특정 구분 기호를 비활성화하려면 `dangerous_patterns` 목록에 추가하세요.\n\n## 특허\n\nMIT","Korean","ko-KR",{"_29":62,"_33":66,"_38":67},{"_31":63,"_33":64,"_35":65},"允许AI助手通过具有全面安全功能的受控界面安全地执行常见的Unix / macOS终端命令。","命令行 MCP 服务器","# 命令行 MCP 服务器\n\n[![PyPI 版本](https://badge.fury.io/py/cmd-line-mcp.svg)](https://badge.fury.io/py/cmd-line-mcp) [![Python 版本](https://img.shields.io/pypi/pyversions/cmd-line-mcp.svg)](https://pypi.org/project/cmd-line-mcp/) [![许可证：MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n\n安全的模型控制协议 (MCP) 服务器，允许 AI 助手以受控的目录访问和命令权限执行终端命令。\n\n## 概述\n\n命令行 MCP 在 AI 助手和终端之间提供了一个安全层。它实现了双重安全模型：\n\n1. **命令权限**：命令分为读取（安全）、写入（更改数据）或系统（影响系统状态），具有不同的批准要求\n2. **目录权限**：命令只能访问明确列入白名单的目录或会话期间批准的目录\n\nAI 助手使用标准化 MCP 工具与此服务器交互，从而实现安全的终端命令执行，同时阻止访问敏感文件或执行危险操作。您可以根据需要将安全级别从严格限制配置为更宽松。\n\n## 主要特点\n\n| 安全 | 可用性 | 一体化 |\n| ------- | ------------ | ------------ |\n| 目录白名单 | 命令分类（读/写/系统） | Claude 桌面兼容性 |\n| 命令过滤 | 持久会话权限 | 标准MCP协议 |\n| 模式匹配 | 命令链（管道等） | 自动批准选项 |\n| 危险的命令阻塞 | 直观的审批工作流程 | 多种配置方法 |\n\n## 支持的命令（开箱即用）\n\n### 读取命令\n\n* `ls` 、 `pwd` 、 `cat` 、 `less` 、 `head` 、 `tail` 、 `grep` 、 `find` 、 `which` 、 `du` 、 `df` 、 `file` 、 `sort`等。\n\n### 写入命令\n\n* `cp` 、 `mv` 、 `rm` 、 `mkdir` 、 `rmdir` 、 `touch` 、 `chmod` 、 `chown`等。\n\n### 系统命令\n\n* `ps` 、 `top` 、 `htop` 、 `who` 、 `netstat` 、 `ifconfig` 、 `ping`等。\n\n## 安全架构\n\n该系统实施多层安全方法：\n\n```\n┌───────────────────────────────────────────────────────────────┐\n│ COMMAND-LINE MCP SERVER │\n├──────────────────┬────────────────────────┬───────────────────┤\n│ COMMAND SECURITY │ DIRECTORY SECURITY │ SESSION SECURITY │\n├──────────────────┼────────────────────────┼───────────────────┤\n│ ✓ Read commands │ ✓ Directory whitelist │ ✓ Session IDs │\n│ ✓ Write commands │ ✓ Runtime approvals │ ✓ Persistent │\n│ ✓ System commands│ ✓ Path validation │ permissions │\n│ ✓ Blocked list │ ✓ Home dir expansion │ ✓ Auto timeouts │\n│ ✓ Pattern filters│ ✓ Subdirectory check │ ✓ Desktop mode │\n└──────────────────┴────────────────────────┴───────────────────┘\n```\n\n所有安全功能都可以根据您的威胁模型和便利性要求从限制性配置为许可性。\n\n## 快速入门\n\n```bash\n# Install\ngit clone https://github.com/yourusername/cmd-line-mcp.git\ncd cmd-line-mcp\npython -m venv venv\nsource venv/bin/activate\npip install -e .\ncp config.json.example config.json\n\n# Run\ncmd-line-mcp # With default config\ncmd-line-mcp --config config.json # With specific config\n```\n\n### 配置选项\n\n服务器支持四种配置方法，按优先顺序排列：\n\n1. **内置默认配置**（default\\_config.json）\n\n2. **JSON配置文件**（建议定制）\n\n ```bash\n cmd-line-mcp --config config.json\n ```\n\n3. **环境变量**（用于特定覆盖）\n\n ```bash\n export CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"~,/tmp\"\n ```\n\n4. **.env 文件**（用于特定环境的设置）\n\n ```bash\n cmd-line-mcp --config config.json --env .env\n ```\n\n默认配置存储在`default_config.json`中，并包含在软件包中。您可以复制此文件来创建自己的自定义配置。\n\n#### 核心配置设置\n\n```json\n{\n \"security\": {\n \"whitelisted_directories\": [\"/home\", \"/tmp\", \"~\"],\n \"auto_approve_directories_in_desktop_mode\": false, \n \"require_session_id\": false,\n \"allow_command_separators\": true\n },\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\"], \n \"write\": [\"touch\", \"mkdir\", \"rm\"],\n \"system\": [\"ps\", \"ping\"]\n }\n}\n```\n\n#### 环境变量格式\n\n环境变量使用可预测的命名模式：\n\n```\nCMD_LINE_MCP_

_\n```\n\n例子：\n\n```bash\n# Security settings\nexport CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES=\"/projects,/var/data\"\nexport CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE=true\n\n# Command additions (these merge with defaults)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq,wc\"\n```\n\n### Claude 桌面集成\n\n#### 设置\n\n1. 安装[Claude 桌面版](https://claude.ai/download)\n2. 在`~/Library/Application Support/Claude/claude_desktop_config.json`中配置：\n\n```json\n{\n \"mcpServers\": {\n \"cmd-line\": {\n \"command\": \"/path/to/venv/bin/cmd-line-mcp\",\n \"args\": [\"--config\", \"/path/to/config.json\"],\n \"env\": {\n \"CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID\": \"false\",\n \"CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE\": \"true\"\n }\n }\n }\n}\n```\n\n#### 推荐的 Claude 桌面设置\n\n为了获得最佳体验，请配置：\n\n* `require_session_id: false` - 防止审批循环\n* `auto_approve_directories_in_desktop_mode: true` - 可选，方便访问\n* 将常用目录添加到白名单中\n\n配置完成后，重新启动Claude for Desktop。\n\n## AI助手工具\n\n该服务器为AI助手提供了以下MCP工具：\n\n| 工具 | 目的 | 需要批准 |\n| ------------------------- | --------- | -------------- |\n| `execute_command` | 运行任意命令类型 | 是的，用于写入/系统命令 |\n| `execute_read_command` | 运行只读命令 | 仅目录批准 |\n| `approve_directory` | 授予目录访问权限 | N/A - 这是一个审批工具 |\n| `approve_command_type` | 授予命令类别的权限 | N/A - 这是一个审批工具 |\n| `list_directories` | 显示授权目录 | 不 |\n| `list_available_commands` | 显示命令类别 | 不 |\n| `get_command_help` | 获取命令使用指南 | 不 |\n| `get_configuration` | 查看当前设置 | 不 |\n\n### 工具示例\n\n#### 目录管理\n\n```python\n# Check available directories\ndirs = await list_directories(session_id=\"session123\")\nwhitelisted = dirs[\"whitelisted_directories\"]\napproved = dirs[\"session_approved_directories\"]\n\n# Request permission for a directory\nif \"/projects/my-data\" not in whitelisted and \"/projects/my-data\" not in approved:\n result = await approve_directory(\n directory=\"/projects/my-data\", \n session_id=\"session123\"\n )\n```\n\n#### 命令执行\n\n```python\n# Read commands (read permissions enforced)\nresult = await execute_read_command(\"ls -la ~/Documents\")\n\n# Any command type (may require command type approval)\nresult = await execute_command(\n command=\"mkdir -p ~/Projects/new-folder\", \n session_id=\"session123\"\n)\n```\n\n#### 获取配置\n\n```python\n# Check current settings\nconfig = await get_configuration()\nwhitelist = config[\"directory_whitelisting\"][\"whitelisted_directories\"]\n```\n\n## 目录安全系统\n\n服务器将命令执行限制在特定目录中，从而防止访问敏感文件。\n\n### 目录安全模式\n\n系统支持三种安全模式：\n\n| 模式 | 描述 | 最适合 | 配置 |\n| -------- | ----------------------- | ------ | ------------------------------------------------- |\n| **严格的** | 仅允许列入白名单的目录 | 最高安全级别 | `auto_approve_directories_in_desktop_mode: false` |\n| **赞同** | 非白名单目录需要明确批准 | 交互式使用 | 标准客户端的默认行为 |\n| **自动批准** | 自动批准 Claude Desktop 的目录 | 方便 | `auto_approve_directories_in_desktop_mode: true` |\n\n### 白名单目录配置\n\n```json\n\"security\": {\n \"whitelisted_directories\": [\n \"/home\", // System directories\n \"/tmp\",\n \"~\", // User's home\n \"~/Documents\" // Common user directories\n ],\n \"auto_approve_directories_in_desktop_mode\": false // Set to true for convenience\n}\n```\n\n### 目录审批流程\n\n1. 在目录中请求命令\n2. 系统检查：\n * 该目录是否在全局白名单中？→**允许**\n * 目录是否已在此会话中被批准？→**允许**\n * 都不是？→**请求批准**\n3. 批准后，目录将在整个会话期间保持批准状态\n\n### 路径格式支持\n\n* 绝对路径： `/home/user/documents`\n* 主目录： `~` （扩展至用户主目录）\n* 用户子目录： `~/Downloads`\n\n### Claude 桌面集成\n\n服务器为 Claude Desktop 维护一个持久会话，确保目录批准在请求之间持续存在并防止批准循环。\n\n## 命令自定义\n\n系统使用命令分类来控制访问：\n\n| 类别 | 描述 | 示例命令 | 需要批准 |\n| --- | ---- | ---------------- | ----- |\n| 读 | 安全操作 | ls，cat，find | 不 |\n| 写 | 数据修改 | mkdir、rm、touch | 是的 |\n| 系统 | 系统操作 | ps、ping、ifconfig | 是的 |\n| 已阻止 | 危险命令 | sudo、bash、eval | 总是被拒绝 |\n\n### 定制方法\n\n```json\n// In config.json\n{\n \"commands\": {\n \"read\": [\"ls\", \"cat\", \"grep\", \"awk\", \"jq\"],\n \"write\": [\"mkdir\", \"touch\", \"rm\"],\n \"system\": [\"ping\", \"ifconfig\", \"kubectl\"],\n \"blocked\": [\"sudo\", \"bash\", \"eval\"]\n }\n}\n```\n\n**环境变量方法：**\n\n```bash\n# Add to existing lists, not replace (comma-separated)\nexport CMD_LINE_MCP_COMMANDS_READ=\"awk,jq\"\nexport CMD_LINE_MCP_COMMANDS_BLOCKED=\"npm,pip\"\n```\n\nMCP 服务器将这些附加功能与现有命令合并，让您无需重新创建完整的命令列表即可扩展功能。\n\n### 命令链\n\n服务器支持三种命令链接方法：\n\n| 方法 | 象征 | 例子 | 配置设置 | | |\n| -- | --- | ------------------------ | -------------------------------- | ---------- | -------------------------------- |\n| 管道 | \\` | \\` | \\`ls | grep txt\\` | `allow_command_separators: true` |\n| 顺序 | `;` | `mkdir dir; cd dir` | `allow_command_separators: true` | | |\n| 背景 | `&` | `find . -name \"*.log\" &` | `allow_command_separators: true` | | |\n\n链中的所有命令都必须来自受支持的命令列表。安全检查适用于整个链。\n\n**快速配置：**\n\n```json\n\"security\": {\n \"allow_command_separators\": true // Set to false to disable all chaining\n}\n```\n\n要禁用特定的分隔符，请将它们添加到`dangerous_patterns`列表中。\n\n## 执照\n\n麻省理工学院","Chinese","zh-CN","githubUser","locale",{"_71":72,"_73":74,"_13":106,"_113":32,"_114":32,"_115":116,"_119":120,"_142":-5,"_16":143,"_144":145,"_146":147,"_33":34,"_158":133,"_159":-5,"_160":-5,"_161":-5,"_162":-5,"_163":164,"_169":-5,"_154":129,"_170":-5,"_171":-5,"_134":172,"_173":174},"addedAt","2025-04-07T17:02:29.907907152Z","argumentsJsonSchema",{"_75":76,"_103":104,"_80":105},"properties",{"_77":78,"_82":83,"_85":86,"_88":89,"_91":92,"_94":95,"_97":98,"_100":101},"CMD_LINE_MCP_COMMANDS_READ",{"_31":79,"_80":81},"Comma-separated list of read commands to add to the defaults","type","string","CMD_LINE_MCP_COMMANDS_WRITE",{"_31":84,"_80":81},"Comma-separated list of write commands to add to the defaults","CMD_LINE_MCP_COMMANDS_SYSTEM",{"_31":87,"_80":81},"Comma-separated list of system commands to add to the defaults","CMD_LINE_MCP_COMMANDS_BLOCKED",{"_31":90,"_80":81},"Comma-separated list of blocked commands to add to the defaults","CMD_LINE_MCP_SECURITY_REQUIRE_SESSION_ID",{"_31":93,"_80":81},"Whether to require a session ID for commands","CMD_LINE_MCP_SECURITY_WHITELISTED_DIRECTORIES",{"_31":96,"_80":81},"Comma-separated list of directories that can be accessed without approval","CMD_LINE_MCP_SECURITY_ALLOW_COMMAND_SEPARATORS",{"_31":99,"_80":81},"Whether to allow command chaining with separators like |, ;, &","CMD_LINE_MCP_SECURITY_AUTO_APPROVE_DIRECTORIES_IN_DESKTOP_MODE",{"_31":102,"_80":81},"Whether to automatically approve directories when using Claude Desktop","required",[],"object",[107,108,109,110,111,112],"category:shell-access","category:os-automation","environment:local","hosting:local-only","language:python","license_grade:a","descriptionMarkdown","descriptionPlainText","githubProject",{"_35":36,"_117":118},"url","https://github.com/andresthor/cmd-line-mcp","githubRepository",{"_121":122,"_123":124,"_125":126,"_127":128,"_33":129,"_130":131,"_136":137,"_138":139,"_141":137},"defaultBranch","main","fullName","andresthor/cmd-line-mcp","githubTreeId",106104,"language","Python","cmd-line-mcp","owner",{"_132":133,"_134":135},"login","andresthor","uid","nVKK0ssbDV","recentStargazers",0,"spdxLicense",{"_33":140},"MIT License","stargazers","glamaJson",5091,"inspectable",false,"integrations",[148],{"_149":150,"_31":156,"_134":157},"brand",{"_151":152,"_33":153,"_154":155},"logoSvgPath","M24 0v24H0V0h24ZM10.933 15.89H6.84v5.52h4.198v-.93H7.955v-1.503h2.77v-.93h-2.77v-1.224h2.978v-.934Zm2.146 0h-1.084v5.52h1.035v-3.6l2.226 3.6h1.118v-5.52h-1.036v3.686l-2.259-3.687Zm5.117 0h-1.208l1.973 5.52h1.19l1.976-5.52h-1.182l-1.352 4.085-1.397-4.086ZM5.4 19.68H3.72v1.68H5.4v-1.68Z",".ENV","slug","dotenv","Supports .env file configuration for environment-specific settings of the MCP server","k1ewhaketj","namespace","npmPackage","prompts","redditPostForeignId","resources","scores",{"_165":166,"_167":-5,"_168":-5},"license",100,"quality","security","simpleIcon","supportedPlatforms","tools","s6abb0eqxm","updatedAt","2025-04-08T14:00:28.697697522Z","meta",[177,180,182,189,192,194,197,200,203,206,209,211,214,217,219,221,223,225,227],{"_178":179},"title","Command-Line MCP Server | Glama",{"_181":32,"_33":31},"content",{"_183":184,"_185":186,"_187":188},"href","https://glama.ai/mcp/servers/@andresthor/cmd-line-mcp","rel","canonical","tagName","link",{"_181":34,"_190":191},"property","og:title",{"_181":32,"_190":193},"og:description",{"_181":195,"_190":196},"https://glama.ai/logo.png","og:logo",{"_181":198,"_190":199},"https://glama.ai/mcp/servers/@andresthor/cmd-line-mcp/og-image","og:image",{"_181":201,"_190":202},"630","og:image:height",{"_181":204,"_190":205},"1200","og:image:width",{"_181":207,"_190":208},"website","og:type",{"_181":184,"_190":210},"og:url",{"_181":212,"_190":213},"Glama – MCP Hosting Platform","og:site_name",{"_181":215,"_33":216},"summary_large_image","twitter:card",{"_181":212,"_33":218},"twitter:title",{"_181":32,"_33":220},"twitter:description",{"_181":198,"_33":222},"twitter:image",{"_181":201,"_33":224},"twitter:image:height",{"_181":204,"_33":226},"twitter:image:width",{"_228":229},"script:ld+json",{"_230":231,"_232":233},"@context","http://schema.org","@graph",[234],{"_230":235,"_236":237,"_238":239},"https://schema.org","@type","BreadcrumbList","itemListElement",[240,247,251,255],{"_236":241,"_242":243,"_33":244,"_245":246},"ListItem","item","https://glama.ai/","Glama","position",1,{"_236":241,"_242":248,"_33":249,"_245":250},"https://glama.ai/mcp","MCP",2,{"_236":241,"_242":252,"_33":253,"_245":254},"https://glama.ai/mcp/servers","Servers",3,{"_236":241,"_242":184,"_33":34,"_245":256},4,"visitorIsMaintainer","routes/_public/mcp/servers/~namespace/~slug/_pages/reviews/_route",{"_68":-5,"_29":260,"_175":281,"_316":-5},{"_71":72,"_73":261,"_13":272,"_113":32,"_114":32,"_115":273,"_119":274,"_142":-5,"_16":143,"_144":145,"_146":277,"_33":34,"_158":133,"_159":-5,"_160":-5,"_161":-5,"_162":-5,"_163":280,"_169":-5,"_154":129,"_170":-5,"_171":-5,"_134":172,"_173":174},{"_75":262,"_103":271,"_80":105},{"_77":263,"_82":264,"_85":265,"_88":266,"_91":267,"_94":268,"_97":269,"_100":270},{"_31":79,"_80":81},{"_31":84,"_80":81},{"_31":87,"_80":81},{"_31":90,"_80":81},{"_31":93,"_80":81},{"_31":96,"_80":81},{"_31":99,"_80":81},{"_31":102,"_80":81},[],[107,108,109,110,111,112],{"_35":36,"_117":118},{"_121":122,"_123":124,"_125":126,"_127":128,"_33":129,"_130":275,"_136":137,"_138":276,"_141":137},{"_132":133,"_134":135},{"_33":140},[278],{"_149":279,"_31":156,"_134":157},{"_151":152,"_33":153,"_154":155},{"_165":166,"_167":-5,"_168":-5},[282,284,286,288,290,291,292,293,294,295,296,297,298,299,300,301,302,303,304],{"_178":283},"Reviews | Command-Line MCP Server | Glama",{"_181":285,"_33":31},"Read community reviews and feedback for Command-Line MCP Server. Discover user experiences, ratings, strengths, and potential limitations of this MCP implementation from actual users.",{"_183":287,"_185":186,"_187":188},"https://glama.ai/mcp/servers/@andresthor/cmd-line-mcp/reviews",{"_181":289,"_190":191},"Reviews | Command-Line MCP Server",{"_181":285,"_190":193},{"_181":195,"_190":196},{"_181":198,"_190":199},{"_181":201,"_190":202},{"_181":204,"_190":205},{"_181":207,"_190":208},{"_181":287,"_190":210},{"_181":212,"_190":213},{"_181":215,"_33":216},{"_181":212,"_33":218},{"_181":285,"_33":220},{"_181":198,"_33":222},{"_181":201,"_33":224},{"_181":204,"_33":226},{"_228":305},{"_230":231,"_232":306},[307],{"_230":235,"_236":237,"_238":308},[309,310,311,312,313],{"_236":241,"_242":243,"_33":244,"_245":246},{"_236":241,"_242":248,"_33":249,"_245":250},{"_236":241,"_242":252,"_33":253,"_245":254},{"_236":241,"_242":184,"_33":34,"_245":256},{"_236":241,"_242":287,"_33":314,"_245":315},"Reviews",5,"review","actionData","errors"]

Rate & Review

MCP directory API