Vale MCP Server
Integrates Vale prose linting into AI coding assistants, enabling users to check text files for style and grammar issues using Vale's powerful linting engine. Provides automated style feedback with smart configuration discovery and rich formatted results.
README
Vale MCP Server
A Model Context Protocol (MCP) server that integrates Vale prose linting into AI coding assistants like Claude Code and Gemini command-line tool.

Overview
This MCP server provides AI assistants with the ability to check files for style and grammar issues using Vale's powerful linting engine. It automatically discovers Vale configuration files and provides formatted, actionable feedback about writing quality.
Features
- ✅ File linting: Check any text file for style issues with Vale.
- 🔍 Smart config discovery: Automatically finds
.vale.inifiles using Vale's native upward search. - 🎯 Configuration priority: Supports
VALE_CONFIG_PATHenvironment variable for explicit config. - 📊 Rich formatting: Returns markdown-formatted results with emojis and severity grouping.
- 🛡️ Graceful degradation: Provides helpful installation guidance when Vale isn't installed.
- 🐛 Debug mode: Optional debug logging for troubleshooting.
System requirements
- Node.js: Version 22 or higher
- Vale: Version 3.0 or higher
Build from source
# Clone or navigate to the project directory
cd vale-mcp-server
# Build the TypeScript project
make build
Install globally
# Install globally using npm link
make install
This creates a global vale-mcp-server command that you can use from anywhere.
To uninstall:
make uninstall
Using with AI assistants
The Vale MCP server can be integrated with AI coding assistants that support the MCP standard.
Claude Code

Add the Vale MCP server to Claude Code using stdio transport:
claude mcp add --transport stdio vale-mcp -- vale-mcp-server
Verify installation:
# List configured servers
claude mcp list
# Check server status
claude mcp get vale-mcp
In Claude Code:
> /mcp
# Shows available MCP tools including vale_status and check_file
See Claude Code MCP documentation for more details.
Gemini CLI

Add the Vale MCP server to Gemini command-line tool:
gemini mcp add vale vale-mcp-server
Verify installation:
# List configured servers
gemini mcp list
# Check in conversation
/mcp
Refer to the Gemini command-line tool MCP documentation for more details.
GitHub Copilot in VS Code

Add the Vale MCP server using VS Code's Command Palette for a quick, no-config setup:
-
Open the command palette: press
Ctrl+Shift+P(Windows/Linux) orCmd+Shift+P(Mac). -
Run Command: Type
MCP: Add Serverand select it. -
Provide Server Information:
- Name:
vale(or any name you prefer) - Type: Select
stdio - Command:
vale-mcp-server - Arguments: Leave empty or add
--debugfor debug mode
- Name:
-
Choose Scope:
- User configuration: Available globally across all workspaces
- Workspace: Available only in the current workspace
Verify in VS Code:
- Open GitHub Copilot Chat
- Type
/mcpto see available MCP tools - The Vale tools should appear in the list
Usage examples
- "Check docs/guide.md using Vale"
- "Lint this markdown file: @README.md"
- "What style issues does Vale find in the documentation?"
Available tools
The server provides three MCP tools:
vale_status
Check if you have installed Vale and can access it.
Returns:
- Installation status (installed/not installed)
- Vale version (if installed)
- Platform-specific installation instructions (if not installed)
Example usage in AI:
"Is Vale installed on this system?"
vale_sync
Download Vale styles and packages by running vale sync. Use this when you encounter errors about missing styles directories.
Parameters:
config_path(optional): Path to.vale.inifile. If not provided, uses the server's configured path or searches in the current directory.
Returns:
- Success/failure status
- Output from the sync operation
- Helpful error messages if sync fails
When to use:
- When you see E100 errors:
"The path '/path/to/styles' does not exist" - After creating or updating a
.vale.inifile - When adding new packages to your Vale configuration
Example usage in AI:
"Run vale_sync to download the required styles" "The styles directory is missing, can you sync Vale?"
check_file
Lint a file at a specific path against Vale style rules.
Parameters:
path(required): Absolute or relative path to the file
Returns:
- Formatted markdown with issues grouped by severity
- Detailed issue information (line, column, rule, message, suggestion)
- Summary statistics (errors, warnings, suggestions)
Example usage in AI:
"Check the README.md file for style issues"
Command-line options
vale-mcp-server [options]
Options:
--debug, --verbose, -v Enable debug logging
--help, -h Show help message
--version Show version number
Development
Project structure
vale-mcp-server/
├── src/
│ ├── index.ts # Main MCP server
│ ├── vale-runner.ts # Vale execution wrapper
│ ├── config.ts # Configuration management
│ └── types.ts # TypeScript type definitions
├── build/ # Compiled JavaScript (generated)
├── tests/ # Test files and fixtures
├── package.json # Node.js package configuration
├── tsconfig.json # TypeScript compiler configuration
├── Makefile # Build and installation tasks
└── README.md # This file
Building for development
# Install dependencies
npm install
# Build TypeScript
npm run build
# Or use Make
make build
Linting
# TypeScript compilation check
npx tsc --noEmit
# Strict unused code check
npx tsc --noUnusedLocals --noUnusedParameters --noEmit
AI usage disclosure
I've created this project using Claude Code with the Claude 4.5 model.
License
This project is licensed under the MIT License. Refer to the LICENSE file for details.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。