JSON to TOON MCP Server
Converts JSON data to TOON (Token-Oriented Object Notation) format and back, reducing token usage by 30-60% for more efficient LLM applications.
README
JSON to TOON MCP Server
A Model Context Protocol (MCP) server that provides tools for converting JSON data to TOON (Token-Oriented Object Notation) format. TOON is a token-efficient format designed specifically for LLM applications, reducing token usage by 30-60% compared to JSON.
Features
- JSON to TOON Conversion: Convert JSON data to TOON format with customizable options
- TOON to JSON Conversion: Convert TOON format back to JSON
- Token Savings Analysis: Analyze potential token savings before conversion
- MCP Protocol Support: Full MCP server implementation for integration with LLM applications
Quick Start
1. Install the Package
npm install -g json-to-toon-mcp-server
2. Configure Claude Desktop
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"json-to-toon": {
"command": "npx",
"args": ["json-to-toon-mcp-server"]
}
}
}
3. Restart Claude Desktop
Restart Claude Desktop and the MCP server will be available.
Installation
npm install
Usage
Running the Server
npm start
The server runs on stdio and can be connected to any MCP client.
Available Tools
1. convert_json_to_toon
Converts JSON data to TOON format.
Parameters:
json_data(string, required): JSON data to convertoptions(object, optional): Conversion optionsdelimiter(string): Delimiter for tabular arrays (default: ",")indentation(string): Indentation string (default: " ")show_length_markers(boolean): Show length markers [N] for arrays (default: true)
Example:
{
"json_data": "{\"users\": [{\"id\": 1, \"name\": \"Alice\"}]}",
"options": {
"delimiter": ",",
"show_length_markers": true
}
}
2. convert_toon_to_json
Converts TOON format back to JSON.
Parameters:
toon_data(string, required): TOON data to convert
Example:
{
"toon_data": "users[1]{id,name}:\n 1,Alice"
}
3. analyze_token_savings
Analyzes potential token savings when converting JSON to TOON.
Parameters:
json_data(string, required): JSON data to analyze
Example:
{
"json_data": "{\"users\": [{\"id\": 1, \"name\": \"Alice\"}]}}"
}
Example Conversions
JSON to TOON
JSON Input:
{
"users": [
{ "id": 1, "name": "Alice", "role": "admin" },
{ "id": 2, "name": "Bob", "role": "user" }
]
}
TOON Output:
users[2]{id,name,role}:
1,Alice,admin
2,Bob,user
Token Savings
The TOON format typically reduces token usage by 30-60% compared to JSON, making it ideal for LLM applications where token efficiency is important.
Development
Running Tests
npm test
Development Mode
npm run dev
Integration with MCP Clients
This server can be integrated with any MCP-compatible client. The server communicates via stdio, making it suitable for integration with various LLM applications and development tools.
Manual Configuration for Claude Desktop
To add this MCP server to Claude Desktop manually, add the following configuration to your claude_desktop_config.json file.
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
To add this MCP server to Claude Desktop manually, add the following configuration to your claude_desktop_config.json:
{
"mcpServers": {
"json-to-toon": {
"command": "npx",
"args": ["json-to-toon-mcp-server"],
"env": {}
}
}
}
Alternative configuration using local installation:
{
"mcpServers": {
"json-to-toon": {
"command": "node",
"args": ["/path/to/json-to-toon-mcp-server/index.js"],
"env": {}
}
}
}
Configuration for development (from source):
{
"mcpServers": {
"json-to-toon": {
"command": "node",
"args": ["index.js"],
"cwd": "/path/to/json-to-toon-mcp-server"
}
}
}
Configuration Notes
- Replace
/path/to/json-to-toon-mcp-serverwith the actual path to your installation - The
npxmethod requires the package to be installed globally (npm install -g json-to-toon-mcp-server) - The server communicates via stdio, so no additional network configuration is needed
- After adding the configuration, restart Claude Desktop for changes to take effect
TOON Format Benefits
- Token Efficient: 30-60% fewer tokens than JSON
- LLM-Friendly: Structured format that's easy for LLMs to parse
- Lossless: Full bidirectional conversion between JSON and TOON
- Human Readable: Maintains readability while being compact
- Schema Aware: Explicit structure declarations help with validation
License
MIT
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。