writing-agents

Create and sync Claude Code and ULPI agent configurations with consistent formatting.

2|5|Updated Nov 23, 2025
One-click install
npx skills add https://github.com/ulpi-io/togglebox --skill writing-agents
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-agents
Source: https://github.com/ulpi-io/togglebox/tree/main/.claude/skills/writing-agents
Command: npx skills add https://github.com/ulpi-io/togglebox --skill writing-agents

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Agents used across Claude Code and ULPI often diverge in structure and formatting, making maintenance painful and error-prone. This Skill provides a unified approach to authoring and syncing agent configurations, ensuring consistency across formats.

Core Features & Use Cases

  • Defines a concise, cross-format agent blueprint that stays in sync between ULPI YAML and Claude Code Markdown.
  • Guides the creation of concise examples covering common patterns like queueing, logging, validation, service layers, error handling, and testing, without exposing code blocks.
  • Supports a workflow to keep both representations aligned and tested, reducing drift between environments.
  • Use cases include creating new agents, updating existing agents, and syncing agent definitions across Claude Code and ULPI ecosystems.

Quick Start

Review the ULPI and Claude Code templates for paired agent entries. Then fill in YAML frontmatter for ULPI and Markdown for Claude Code to ensure synchronized behavior. Validate that examples in both formats match and that TodoWrite rules are included.

Frequently Asked Questions about writing-agents

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

FAQPage Schema
How do I keep Claude Code and ULPI agent configurations synchronized?

To keep Claude Code and ULPI agent configurations synchronized, use a unified blueprint that pairs ULPI YAML frontmatter with Claude Code Markdown entries, enforcing matching examples and consistent formatting to prevent structural drift across projects.

What is the frontmatter contract for ULPI and Claude Code agent configs?

The frontmatter contract for ULPI and Claude Code agent configs requires a defined name and description paired with 8–10 representative example patterns and TodoWrite rules to ensure consistent behavior across both ecosystems.

How do I standardize agent formatting across multiple projects?

You standardize agent formatting across multiple projects by applying a cross-format agent blueprint that enforces concise paired examples covering common patterns like logging, validation, and error handling, preventing structural divergence.

Does this approach support both new agent creation and updating existing agents?

Yes, the approach supports both new agent creation and updating existing agents by applying a unified configuration structure that enforces paired examples and frontmatter rules, ensuring both operations remain fully synchronized.

What common patterns should agent examples cover?

Agent examples should cover 8–10 representative patterns including queueing, logging, validation, service layers, error handling, and testing, presented concisely without exposing code blocks in the configuration files.

Why do agent definitions diverge between Claude Code and ULPI environments?

Agent definitions diverge between Claude Code and ULPI because ULPI YAML and Claude Code Markdown use different structural formats, causing maintenance drift; a synchronized blueprint with paired entries resolves this by enforcing matching examples.