Large tasks can be hard to finish all at once. This skill helps you break them into smaller steps. You act as the planner and overseer. You do not do the work yourself. Instead you send out sub-agents to do each step. Each sub-agent gets a model that fits its task. A separate judge agent checks every step before you move on. This makes sure each part is done correctly. The judge gives you a pass or fail. If it fails you can try again. This way complex jobs get done reliably step by step.
This method works for tasks like refactoring code or updating many files. You save time because sub-agents work in parallel with the judge. The judge creates a checklist for each step first. This keeps the work consistent and high quality. Anyone who needs to handle big, multi-step projects can benefit from this skill.
Global
mkdir -p ~/.claude/skills/do-in-stepsProject
mkdir -p .claude/skills/do-in-stepsSource 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