use-toon

Delegate agent tasks using TOON with schema.org vocabulary.

3|Updated Nov 28, 2025

Search Tags:#prompt engineering #subagents #schema.org #TOON #agent prompting

Ready to run

npx skills add https://github.com/rayk/lucid-toolkit --skill use-toon

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: use-toon
Source: https://github.com/rayk/lucid-toolkit/tree/main/plugins/luc/skills/use-toon
Command: npx skills add https://github.com/rayk/lucid-toolkit --skill use-toon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use TOON with schema.org vocabulary for prompting and instructing subagents and builtin agents. Use when delegating tasks to agents, structuring agent prompts, or specifying expected response formats. DO NOT use for external API calls or when JSON parsing is required.

Core Features & Use Cases

Clear task structure: schema.org Action types define what the agent should do
Expected output format: Specify result type so agents return consistent data
Token efficiency: ~30-60% fewer tokens than JSON for more reasoning context
Semantic interoperability: Standard vocabulary reduces ambiguity

Quick Start

Example: Instruct a subagent to perform a SearchAction on a codebase and return results in TOON format, then provide an ItemList as the expectedResult.

Frequently Asked Questions about use-toon

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I prompt agents with schema.org vocabulary instead of JSON?▼

TOON uses schema.org Action types and @id/@type fields to structure agent prompts with ~30-60% fewer tokens than JSON. Format task descriptions and expected result types in TOON, then pass to agents or subagents for token-efficient, interoperable responses without external API calls.

What's the difference between TOON and JSON for agent prompting?▼

TOON achieves token efficiency and semantic interoperability using schema.org vocabulary, reducing prompt size by 30-60% compared to JSON while avoiding JSON parsing requirements. TOON works best for agent-to-agent workflows; JSON remains necessary when external APIs require it.

Can I use TOON to define expected output formats for subagents?▼

Yes. TOON supports specifying result schemas with @type and expectedResult fields, allowing subagents to return consistent, structured data. Use simple or tabular arrays within TOON to enforce explicit task descriptions and output formats across agent workflows.

When should I not use TOON for agent instructions?▼

Do not use TOON for external API calls or when JSON parsing is required by downstream systems. TOON is designed for agent-to-agent delegation and internal workflow prompting where schema.org vocabulary and token efficiency provide the most value.

How do I structure a task delegation with TOON and schema.org?▼

Define the agent action using schema.org Action types with @type and @id fields, specify the task in an explicit description, and declare the expectedResult type. TOON formatting ensures consistent, interoperable responses while minimizing token usage in agent-to-agent communication.