ASON MCP Server
Enables compression and decompression of JSON data using the token-optimized ASON format, reducing token usage by 20-60% for LLM applications.
README
ASON MCP Server
Model Context Protocol (MCP) server for ASON compression/decompression. Enables Claude Desktop, Cline, Continue, and other MCP clients to compress JSON data using the token-optimized ASON format.
Features
- compress_json: Convert JSON to ASON format (20-60% token reduction)
- decompress_ason: Convert ASON back to JSON (lossless)
- get_compression_stats: Analyze compression metrics without performing compression
- configure_compressor: Customize compression settings globally
Installation
Option 1: Local Development
cd mcp-server
npm install
npm run build
Option 2: Global Installation
npm install -g @ason-format/mcp-server
Usage with Claude Desktop
Add to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"ason": {
"command": "node",
"args": [
"/absolute/path/to/ason/mcp-server/dist/index.js"
]
}
}
}
Or if installed globally:
{
"mcpServers": {
"ason": {
"command": "ason-mcp"
}
}
}
Available Tools
1. compress_json
Compress JSON data to ASON format.
Input:
{
"json": {"users": [{"id": 1, "name": "Alice"}, {"id": 2, "name": "Bob"}]},
"config": {
"indent": 1,
"delimiter": "|",
"useReferences": true,
"useSections": true,
"useTabular": true
}
}
Output:
ASON 2.0 Output:
users:[2]{id,name}
1|Alice
2|Bob
Configuration used:
{
"indent": 1,
"delimiter": "|",
"useReferences": true,
"useSections": true,
"useTabular": true
}
2. decompress_ason
Decompress ASON back to JSON.
Input:
{
"ason": "users:[2]{id,name}\n1|Alice\n2|Bob"
}
Output:
{
"users": [
{"id": 1, "name": "Alice"},
{"id": 2, "name": "Bob"}
]
}
3. get_compression_stats
Analyze compression statistics.
Input:
{
"json": {"users": [{"id": 1, "name": "Alice"}, {"id": 2, "name": "Bob"}]},
"config": {
"indent": 1,
"delimiter": "|"
}
}
Output:
Compression Statistics:
Original Tokens: 32
Compressed Tokens: 15
Reduction: 53.13%
Original Size: 78 bytes
Compressed Size: 41 bytes
Savings: 37 bytes
Configuration:
{
"indent": 1,
"delimiter": "|",
"useReferences": true,
"useSections": true,
"useTabular": true
}
4. configure_compressor
Update global compression settings.
Input:
{
"config": {
"indent": 2,
"delimiter": ",",
"useReferences": false
}
}
Output:
Global configuration updated:
{
"indent": 2,
"delimiter": ",",
"useReferences": false,
"useSections": true,
"useTabular": true
}
Configuration Options
| Option | Type | Default | Description |
|---|---|---|---|
indent |
number | 1 |
Indentation level for nested structures |
delimiter |
string | `" | "` |
useReferences |
boolean | true |
Enable $var reference deduplication |
useSections |
boolean | true |
Enable @section organization for objects |
useTabular |
boolean | true |
Enable [N]{fields} tabular format for arrays |
minFieldsForSection |
number | 3 |
Minimum fields to create a @section |
minRowsForTabular |
number | 2 |
Minimum rows for tabular array format |
minReferenceOccurrences |
number | 2 |
Minimum occurrences to create a $var reference |
Development
# Build
npm run build
# Watch mode
npm run watch
# Test locally
node dist/index.js
Example Usage in Claude Desktop
User: Can you compress this JSON for me?
{
"products": [
{"id": 1, "name": "Laptop", "price": 999},
{"id": 2, "name": "Mouse", "price": 25}
]
}
Claude: I'll use the compress_json tool to compress this JSON.
[Uses MCP tool compress_json]
Result:
products:[2]{id,name,price}
1|Laptop|999
2|Mouse|25
This compressed version uses 45% fewer tokens!
๐ Publishing
To release a new version:
# Run the release script
./scripts/release.sh
# 1. Select version bump (patch/minor/major)
# 2. Update CHANGELOG.md when prompted
# 3. Confirm push
# GitHub Actions will automatically:
# - Build the package
# - Publish to NPM (@ason-format/mcp-server)
# - Create GitHub Release
๐งช Supported MCP Clients
- โ Claude Desktop (Anthropic)
- โ Cline (VS Code extension)
- โ Continue (VS Code extension)
- โ Any MCP client with stdio transport
๐ What is ASON 2.0?
ASON (Aliased Serialization Object Notation) 2.0 is a token-optimized JSON compression format designed for LLMs. It reduces token usage by 20-60% while maintaining 100% lossless round-trip fidelity.
Key features:
- Sections (
@section) - Organize related objects - Tabular Arrays (
key:[N]{fields}) - CSV-like format for uniform arrays - References (
$var) - Deduplicate repeated values - Pipe Delimiter (
|) - More token-efficient than commas
Learn more: ason-format.github.io/ason
๐ Documentation
- Changelog - Version history
- ASON Format - Core library
- Model Context Protocol - MCP specification
- Claude Desktop - Download Claude Desktop
๐ License
MIT ยฉ ASON Project Contributors
๐ค Contributing
Contributions welcome! Please open an issue or pull request.
๐ Links
- NPM: https://www.npmjs.com/package/@ason-format/mcp-server
- GitHub: https://github.com/ason-format/mcp-server
- Issues: https://github.com/ason-format/mcp-server/issues
- ASON Core: https://github.com/ason-format/ason
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.