Can I use TOON for external application APIs?

No, TOON is specifically optimized for internal AI agent communication, memory layers, and orchestration within the plan-marshall framework, rather than general-purpose web APIs.

How much can TOON reduce token usage?

For uniform arrays and repetitive data structures, TOON can achieve a 30% to 60% reduction in tokens compared to standard JSON or YAML.

When should I activate the TOON format skill?

Activate this skill when designing agent handoff templates, optimizing memory persistence layers, or converting large JSON fixtures into token-efficient representations.

What is a Pattern 10 skill in Claude Code?

Pattern 10 refers to a Reference Library skill, which acts as a pure knowledge base that Claude can query on-demand for specifications without requiring execution logic.

What is the TOON format?

TOON (Token-Oriented Object Notation) is a compact, human-readable data format designed to minimize token usage for AI models while maintaining structural clarity.

TOON Format Reference

Name: TOON Format Reference
Author: cuioss

bycuioss

0•

Productivity & Workflow

Optimizes AI agent communication and memory persistence by implementing the token-efficient Token-Oriented Object Notation (TOON) format.

The TOON Format Reference skill provides a comprehensive technical guide and usage patterns for Token-Oriented Object Notation, a compact data structure designed to reduce token usage by up to 60% compared to JSON. Specifically tailored for the plan-marshall orchestration layer, this skill enables Claude to handle high-density data like agent handoffs, structured memory persistence, and tabular test fixtures with significantly lower overhead. It offers specific guidance on syntax conversion, field declarations, and inter-agent data exchange protocols to ensure reliable and predictable AI outputs during complex multi-step coding workflows.

Key Features

01Technical TOON syntax and semantics reference guide

02Tabular data representation patterns for high-density reports

03Memory persistence strategies for internal state management

04Token efficiency (30-60% reduction vs JSON for uniform arrays)

05Standardized agent-to-agent handoff templates

060 GitHub stars

Use Cases

01Optimizing the transfer of large issue lists or coverage reports between agents

02Reducing context window usage during multi-agent collaboration chains

03Standardizing internal memory storage for long-running AI development tasks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cuioss/plan-marshall ref-toon-format

For use in Claude.ai and ChatGPT

Download Skill