Do And Judge
neolabhq/context-engineering-kitAn AI agent can handle a single task by using two helpers. One helper does the work. Another helper acts as a judge to check if the work is good. This pattern helps catch mistakes without the main agent looking at the code.
The judge uses a set of rules made by a meta-judge. If the work does not pass, the agent tries again with feedback. It can retry up to two times. This creates a quality gate that stops bad results from moving forward.
This method works well for tasks like refactoring code or writing new features. It gives a fresh perspective from an independent agent. The whole process runs faster by doing the judge rules and the work at the same time.
Add Do And Judge skill to your workflow
Global
mkdir -p ~/.claude/skills/do-and-judgeProject
mkdir -p .claude/skills/do-and-judgeSource Repository
More AI & Agent Building Claude Code Skills
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