judge-with-debate — Claude Code skill: scoring by debate
The skill runs three independent judges who argue over a solution's score until they agree. Useful when a single pass feels too thin for a real verdict.
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a claude-codeWhat judge-with-debate does
The skill implements the Multi-Agent Debate pattern. A meta-judge runs once and writes an evaluation specification: rubrics, checklists and a scoring scale. Then three judges analyse the solution in parallel, each on their own, and save reports to .specs/reports/{solution-name}-{YYYY-MM-DD}.[1|2|3].md.
Debate rounds follow, up to three of them. Each judge reads the other reports, looks for score gaps wider than one point on any criterion, defends their rating with evidence from the solution and the specification, and revises when the counter-arguments hold. Judges talk only through the filesystem — the orchestrator stays out of it, otherwise its context overflows.
The reports directory .specs/reports has to exist before the run; the skill creates it with mkdir -p .specs/reports. Both the meta-judge and the judges run on Opus, so expect a heavier bill than a plain review. On a small file the debate mostly adds latency.
What it helps with
- score a solution against a rubric instead of a gut call
- cut the bias that comes with one reviewer
- surface the disputed criteria and both sides of the argument
- get a report with per-criterion scores and quotes from the code
- compare several solutions against one specification
Example prompts for the agent
- evaluate
src/api/users.tswith judge-with-debate - run the judge debate on this module and show where they disagree
- do a multi-agent debate on this file and save the reports to
.specs/reports
How to install judge-with-debate
1Skills CLI (Vercel)
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a claude-code2Manually: put the files in place
~/.claude/skills/judge-with-debate/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/judge-with-debate/
1Skills CLI (Vercel)
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a codex2Manually: put the files in place
.agents/skills/judge-with-debate/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a cursor2Manually: put the files in place
.cursor/skills/judge-with-debate/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install NeoLabHQ/context-engineering-kit judge-with-debate2Manually: put the files in place
.github/skills/judge-with-debate/SKILL.md1Skills CLI (Vercel)
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a gemini-cli2Platform CLI
gemini skills install https://github.com/NeoLabHQ/context-engineering-kit --path skills/judge-with-debateManually: put the files in place
.devin/skills/judge-with-debate/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about judge-with-debate
How do I install judge-with-debate in Claude?
Run npx skills add NeoLabHQ/context-engineering-kit --skill judge-with-debate -a claude-code. The skill file then lands at ~/.claude/skills/judge-with-debate/SKILL.md.
What does judge-with-debate do?
It puts a solution through three independent judges, then through debate rounds until consensus or three rounds are up. You get reports with per-criterion scores and the reasoning behind them.
Does the skill need anything set up first?
The .specs/reports directory must exist before the run, and the skill creates it itself. The meta-judge and the judges run on Opus.
