skill-builder

Guides authoring, structuring, and validating custom SKILL.md files for agent skills.

1|Updated Jul 2, 2026
One-click install
npx skills add https://github.com/tarabakz25/my-skills --skill skill-builder-tarabakz25
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-builder
Source: https://github.com/tarabakz25/my-skills/tree/main/skill-builder
Command: npx skills add https://github.com/tarabakz25/my-skills --skill skill-builder-tarabakz25

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Agents lack durable procedural memory across sessions, so recurring tasks get re-solved from scratch each time. This Skill defines the conventions, frontmatter format, and quality checklist needed to write reusable SKILL.md files that agents can load on demand. ## Core Features & Use Cases - Canonical Skill Format: Defines the required YAML frontmatter fields (name, description, version, author, license, platforms, metadata.hermes) and the standard body sections (Overview, When to Use, Pitfalls, Verification Checklist). - Creation Workflow: Provides a six-step process from identifying a trigger through drafting, validating frontmatter with a Python snippet, and committing to version control. - Decision Guidance: Clarifies when content belongs in a skill versus agent memory versus a standalone script. - Use Case: You keep repeating the same deployment procedure across sessions. Use this Skill to author a deploy-app/SKILL.md with exact commands, pitfalls, and verification steps so future sessions execute it consistently. ## Quick Start Ask the agent to create a new skill for a recurring task, following the skill-builder conventions for frontmatter and body structure.

Frequently Asked Questions about skill-builder

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

FAQPage Schema
How do I create a custom skill for an AI agent?▼

Create a directory under ~/.skills/ with a SKILL.md file containing YAML frontmatter (name, description, version, author, license, platforms) followed by body sections: Overview, When to Use, workflow steps, Common Pitfalls, and a Verification Checklist.

What frontmatter fields are required in a SKILL.md file?▼

Required fields are name (lowercase kebab-case, max 64 chars), description (max 1024 chars), version, author, license, and platforms. The metadata.hermes section with tags and related_skills is strongly recommended.

When should I use a skill versus agent memory or a script?▼

Use a skill for reusable workflows requiring reasoning, memory for durable facts like user preferences, and scripts for deterministic automation with no agent judgment. If a human would read instructions and think, it is a skill.

How do I validate SKILL.md frontmatter?▼

Parse the file with Python's yaml module after confirming it starts with ---, then assert the name and description fields exist and stay within the 64 and 1024 character limits. Total content should stay under 100,000 characters.

What are common mistakes when writing agent skills?▼

Common mistakes include overly generic triggers, skipping the pitfalls section, hardcoding unstable paths, exceeding 20,000 characters without splitting into references/, omitting verification steps, and duplicating user preferences that belong in memory.