SpellChecker MCP Server
Provides fast, multilingual spell-checking for text and files, featuring syntax-aware parsing that intelligently checks only comments and strings in code. It supports over 15 languages and enables directory-wide scanning with custom dictionary management.
README
SpellChecker MCP Server
A fast, multilingual Model Context Protocol (MCP) server that provides spell-checking capabilities to Large Language Models. This server enables LLMs to check spelling in text, files, and entire projects with syntax-aware parsing for code files.
Features
- Multi-language support: 15+ languages including English, Spanish, French, German, Portuguese, Italian, Dutch, Polish, Russian, Ukrainian, Swedish, Danish, and Norwegian
- Syntax-aware checking: Intelligently checks only comments, strings, and documentation in code files
- File and folder scanning: Check individual files or entire project directories
- Modular language activation: Enable only the languages you need
- Fast spell checking: Powered by nspell for efficient processing
- Custom dictionary management: Add words to personal dictionaries
- Smart code parsing: Understands HTML, JavaScript, TypeScript, Python, and more
- MCP-compliant: Works with any MCP-compatible client
Installation
As an MCP Server
- Clone the repository:
git clone https://github.com/yourusername/spellchecker-mcp-server.git
cd spellchecker-mcp-server
- Install dependencies and build:
yarn install
yarn build
The post-install script will automatically download the required dictionary files.
Using with Claude Desktop
Add the server to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"spellchecker": {
"command": "node",
"args": ["/path/to/spellchecker-mcp-server/dist/index.js"],
"env": {
"SPELLCHECKER_LANGUAGES": "en-US,es,fr"
}
}
}
}
Configuring Languages
You can configure which languages to enable in three ways:
-
Environment variable (recommended for Claude Desktop):
SPELLCHECKER_LANGUAGES="en-US,es,fr" node dist/index.js -
Configuration file:
SPELLCHECKER_CONFIG="/path/to/spellchecker.config.json" node dist/index.js -
Edit the default in
src/config.tsbefore building
Using with Other MCP Clients
The server runs on stdio and can be integrated with any MCP-compatible client:
node /path/to/spellchecker-mcp-server/dist/index.js
Available Tools
check_spelling
Checks text for spelling errors and returns misspellings with suggestions.
Parameters:
text(required): The text to checklanguage(optional): Language code (default: "en-US")includeLineNumbers(optional): Include line/column positions (default: false)
Example:
{
"text": "This is a tset of the speling checker",
"language": "en-US",
"includeLineNumbers": true
}
is_correct
Checks if a single word is spelled correctly.
Parameters:
word(required): The word to checklanguage(optional): Language code (default: "en-US")
get_suggestions
Gets spelling suggestions for a word.
Parameters:
word(required): The word to get suggestions forlanguage(optional): Language code (default: "en-US")limit(optional): Maximum suggestions to return (default: 5)
add_to_dictionary
Adds a word to the personal dictionary for a language.
Parameters:
word(required): The word to addlanguage(optional): Language code (default: "en-US")
list_languages
Lists all available languages for spell checking.
check_file
Checks spelling in a single file with syntax-aware parsing.
Parameters:
filePath(required): Path to the file to checklanguage(optional): Language code (default: "en-US")syntaxAware(optional): Enable syntax-aware parsing (default: true)
Example:
{
"filePath": "/path/to/app.tsx",
"language": "en-US",
"syntaxAware": true
}
check_folder
Checks spelling in all files within a folder.
Parameters:
folderPath(required): Path to the folder to checklanguage(optional): Language code (default: "en-US")recursive(optional): Check subfolders (default: true)fileTypes(optional): Array of file extensions to checksyntaxAware(optional): Enable syntax-aware parsing (default: true)
Example:
{
"folderPath": "/path/to/project",
"language": "en-US",
"recursive": true,
"fileTypes": [".js", ".tsx", ".md"],
"syntaxAware": true
}
Supported Languages
The server supports 15+ languages. Enable only what you need:
en-US- English (United States)en-GB- English (United Kingdom)es- Spanishfr- Frenchde- Germanpt- Portuguesept-BR- Portuguese (Brazil)it- Italiannl- Dutchpl- Polishru- Russianuk- Ukrainiansv- Swedishda- Danishnb- Norwegian Bokmål
Syntax-Aware Features
When syntaxAware is enabled, the spell checker intelligently parses:
- Comments: Single-line and multi-line comments in all major languages
- String literals: Only checks strings that look like human-readable text
- JSX/TSX content: Text between JSX tags
- Documentation: Docstrings, JSDoc comments, etc.
- Markdown: Ignores code blocks and inline code
- HTML/XML: Checks text content and comments
The checker automatically ignores:
- Variable names and identifiers
- Programming keywords
- URLs and file paths
- Hex colors and numbers
- Code within backticks in Markdown
Development
Prerequisites
- Node.js >= 16.0.0
- Yarn package manager
Setup
# Install dependencies
yarn install
# Run in development mode
yarn dev
# Build for production
yarn build
# Type check
yarn typecheck
Using Just Commands
If you have just installed:
# Show available commands
just
# Fresh install
just fresh
# Run development server
just dev
# Update dictionaries
just dictionaries
Architecture
The server uses:
- @modelcontextprotocol/sdk: MCP protocol implementation
- nspell: Hunspell-compatible spell checker
- glob: File pattern matching for directory scanning
- TypeScript: Type-safe development
- Dictionary files: From the wooorm/dictionaries project
Configuration File
Create a spellchecker.config.json file:
{
"languages": ["en-US", "es", "fr"],
"defaultLanguage": "en-US",
"scanOptions": {
"syntaxAware": true,
"recursive": true,
"fileTypes": [
".txt", ".md", ".mdx",
".js", ".jsx", ".ts", ".tsx",
".py", ".java", ".go"
],
"ignorePaths": [
"node_modules",
".git",
"dist",
"build"
]
}
}
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Acknowledgments
- Dictionary files from wooorm/dictionaries
- Built on the Model Context Protocol
Troubleshooting
Dictionaries not loading
If you see "No dictionaries loaded" error:
yarn run postinstall
# or
just dictionaries
Server not starting
Ensure you've built the project:
yarn build
Language not available
Check available languages with the list_languages tool or ensure the dictionary files exist in the dictionaries/ folder.
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。