skills.rest
diegocondo10
diegocondo10COMMUNITY
Explore all skills by this author →
GitHub

writing-skills

Design and validate reusable SKILL.md documentation for AI agents.

Updated Aug 27, 2026
Search Tags:#tdd#documentation#prompt engineering#verification#subagents#skill authoring#red-green-refactor
One-click install
npx skills add https://github.com/diegocondo10/sinai-intranet-backend --skill writing-skills-diegocondo10
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/diegocondo10/sinai-intranet-backend/tree/main/.agents/skills/writing-skills
Command: npx skills add https://github.com/diegocondo10/sinai-intranet-backend --skill writing-skills-diegocondo10

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you create and revise agent skills that are easy to discover, concise to load, and reliable under pressure instead of leaving behavior to vague or untested documentation.

Core Features & Use Cases

  • Skill design: Shape clear SKILL.md frontmatter, naming, and structure so the skill is easy to find and activate.
  • Pressure testing: Check whether the instructions still work when the model faces time pressure, sunk cost, or rationalizations.
  • Loophole closure: Tighten wording to prevent common failures such as skipping verification, drifting from the rule, or relying on unstated assumptions.
  • Reference organization: Keep heavier guidance in supporting references or examples when the main skill should stay compact.
  • Use case: Use it when creating a new skill, editing an existing one, or validating that an agent follows the guidance before deployment.

Quick Start

Ask the AI to draft or revise a skill for your target workflow, then test it under realistic pressure and refine the wording until the behavior is consistent.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I write reliable AI agent documentation that doesn't fail under pressure?▼

Skill authoring creates reusable process documentation with YAML frontmatter, pressure-tested instructions, and compliance tests to ensure AI agents behave reliably and avoid loopholes during execution.

What is the best way to structure a SKILL.md file for AI prompt engineering?▼

Structure SKILL.md with required YAML frontmatter name and description, concise operational guidance, supporting references or scripts only when needed, and verification tests to confirm compliance before deployment.

How do I pressure test AI agent instructions to close loopholes?▼

Pressure test agent instructions by simulating time pressure, sunk cost, and rationalizations, then tighten wording to close loopholes like skipping verification, drifting from rules, or relying on unstated assumptions.

Do I need to write tests for AI agent skills before deployment?▼

Yes, you must write tests confirming compliance before deployment to ensure the agent follows guidance consistently and avoids behavior failures caused by vague or untested documentation.

Can I use TDD principles to create and validate subagent skills?▼

Yes, you can apply red-green-refactor TDD principles to skill authoring by writing compliance tests first, then drafting and refining SKILL.md documentation until the agent behavior passes all verification tests.

Related Skills

Explore More Software Engineering Skills→
different-aidifferent-ai
249

skill-authoring

Define OpenCode skills with SKILL.md frontmatter and validate discoverability.

Official
Basic
devtitusdevtitus

writing-skills

Automate skill documentation creation and validation using the RED-GREEN-REFACTOR workflow.

Community
Intermediate
Nuwanda04Nuwanda04

senior-prompt-engineer

Analyze prompts, evaluate RAG systems, and design agentic workflows.

Community
Advanced
xuithoixuithoi

writing-skills

Apply Test-Driven Development principles to author and maintain agent skills.

Community
Advanced