Token-Oriented Object Notation (TOON) – compact, human-readable serialization of JSON data for LLM prompts

Lightweight MCP orchestrator with TOON compression (30-90% token savings) and lazy loading for efficient AI agent workflows

Node.js compression middleware

Lightweight MCP orchestrator with TOON compression (30-90% token savings) and lazy loading for efficient AI agent workflows

Cost-optimized model routing for Oh-My-Pi — routes prompts to cheap/mid/expensive models based on task complexity. Tracks per-turn and session costs. Uses TOON compression for conversation history optimization.

Avatar style for DiceBear

Token optimization middleware for AI APIs — reduce costs by 70-91% with TOON compression, cache alignment, and caveman mode

Token-efficient JSON-to-TOON compression middleware and CLI for AI applications, with Gemini optional for AI analysis.

CLI for JSON ↔ TOON conversion using @toon-format/toon

High performance (de)compression in an 8kB package

Encode JSON → TOON (40-60% token savings for LLMs!)

TypeScript definitions for compression

A Model Context Protocol (MCP) server for interacting with Microsoft 365 and Office services through the Graph API

Prepare compressed versions of assets to serve them with Content-Encoding

Middleware and Plugin for request compression.

Compress images in the browser

TOON v4.1 encoder and decoder, plus TOONL v0.2 append-only streaming. Dependency-free ESM, conformant against the official spec corpus.

ECC JS code based on JSBN

YVON Engine — AI Agent OS kernel. 13 agents, Context Intelligence Engine, TOON compression, knowledge graphs, Hermes sync. One npm install, full agent team.

The Continuity Layer for AI Agents — an MCP memory server that preserves project knowledge, decisions, and conventions across sessions so every session starts where the last one ended. Offline, token-efficient, no LLM, no server; works with 20+ agents.