Score test-kitchen implementations using a fixed five-criteria rubric.
npx skills add https://github.com/2389-research/claude-plugins --skill judge
Please help me install this Agent Skill. Skill: judge Source: https://github.com/2389-research/claude-plugins/tree/main/test-kitchen/skills/judge Command: npx skills add https://github.com/2389-research/claude-plugins --skill judge
Scores test-kitchen implementations using a fixed five-criteria rubric to provide objective, comparable assessments across cookoff and omakase-off events.
Run the judge on your candidate implementations to produce a complete scoring report in the prescribed format.
High-intent search queries and answers about installing and using this skill.
Score multiple implementations by applying a fixed five-criteria rubric to generate objective, comparable assessments across cookoff and omakase-off events. This structured approach enforces consistent evaluation through Gate Check and scoring worksheet sections.
A rubric-based scoring worksheet is a structured format for evaluating test-kitchen implementations using fixed criteria including Gate Check, Feasibility, Scoring Worksheet, Judge Scorecard, and Hard Gates sections to ensure objective comparisons across candidates.
Generate a comparable scoring report by running the judge on candidate implementations like impl-1, impl-2, impl-3 or variant-a, variant-b to produce structured Gate Check and scoring worksheet outputs suitable for decision-making.
Yes, the fixed five-criteria rubric applies to both cookoff and omakase-off scenarios to compare multiple implementations fairly. It enforces the required output format, structured worksheets, and rubric-based scoring rules for objective assessment.
Required sections include Gate Check, Feasibility, Scoring Worksheet, Judge Scorecard, and Hard Gates. These sections enforce consistent evaluation and produce a complete scoring report in the prescribed format for comparing candidate implementations.
ParamchoudharyGenerate STAR stories, interview questions, and talking points from a resume.
pedapudiAudit zicato evaluation boards for harness correctness using canonical artifacts.
FVossebeldEvaluate agent outputs with weighted rubric criteria and per-dimension scores.
Victory-HugoEvaluate scientific literature methodology, biases, and evidence quality.
