Making a new skill for an AI agent is like writing a test first. You create a test case using a subagent and watch it fail. Then you write the skill documentation to fix the problem. This approach is called test-driven development for skills.
The skill helps you teach an AI agent new abilities step by step. You see what the agent does wrong without the skill. Then you write clear instructions to make it right. You keep testing and improving until the agent always does the right thing.
Anyone who wants to build better AI agents can use this guide. It turns general AI helpers into specialized experts for specific tasks.
Global
mkdir -p ~/.claude/skills/create-skillProject
mkdir -p .claude/skills/create-skillSource Repository
Find Skillsvercel-labs/skills
Find and install the perfect skill to extend your AI agent
Microsoft Foundrymicrosoft/azure-skills
Build, deploy, and improve AI agents on Microsoft Foundry from start to finish
Azure Aimicrosoft/azure-skills
Search, transcribe, and analyze with Azure AI tools for smarter apps
Azure Hosted Copilot Sdkmicrosoft/azure-skills
Build, deploy, and manage your Copilot SDK apps on Azure with ease
Triagemattpocock/skills
Triage issues with a state machine driven by clear roles and agent briefs
Handoffmattpocock/skills
Hand off your work to another AI agent with a clear summary
Image Editagentspace-so/runcomfy-agent-skills
Smart router picks the best AI model for your image editing needs
Agentspaceagentspace-so/skills
See your AI agent's live folder from any browser instantly