AI Agent Skills: Everything You Need to Know
In 2024, AI coding agents (Claude, Cursor, Copilot, etc.) became capable enough to do real work. Teams started writing long instruction files and system prompts. But there was no standard way to package, share, or discover them. Every team reinvented the wheel. Today, we’re creating a complete handbook on how the solution to the issue above came about, how to use it, and how it became one of the…
In 2024, AI coding agents like Claude, Cursor, and Copilot became proficient enough to perform real work. However, there was no standardized method for packaging, sharing, or discovering these agents. To address this, a complete handbook on Agent Skills and the skills.sh platform was created, which simplifies the usage of these agents.
Agent Skills are a lightweight, open format that provide AI agents with specialized knowledge and workflows. They consist of a folder containing a SKILL.md file with YAML frontmatter and markdown instructions. Skills.sh is an open directory and CLI for the Agent Skills ecosystem, functioning as the npm for AI agent capabilities.
It supports various agents including Claude Code, Cursor, Codex, GitHub Copilot, Windsurf, and Gemini CLI. Skills are installed using the command npx skills add owner/repo. There are four examples of skills, including a universal PR Review skill, a Flutter Widget Test skill, a Generate Changelog skill, and a Brand Voice skill. The process of creating a skill involves creating a folder, writing the frontmatter with the name and description, providing clear instructions, and testing with real prompts.
Skills can be shared within a team or publicly on skills.sh, either by committing the skills folder to a repository or by pushing a public GitHub repository containing the skill.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
