Output Quality Rubrics Skill for ChatGPT: Calibrate Accuracy and Usefulness
Break accuracy, relevance, completeness, and usefulness into observable scoring criteria. Define levels and check whether independent reviewers score the same examples consistently.
中文Install in ChatGPT
Use an eligible ChatGPT workspace with permission to create skills and, for uploads, upload permission. The installation path for this text-only adaptation was checked against OpenAI documentation. Workspace runtime verification is pending.
- Download claude-plugin/evaluation/skills/output-quality-rubrics/SKILL.md from the pinned GitHub source. Keep the name, description, and instructions for output-quality-rubrics together as a standalone text skill.
- In a ChatGPT workspace with skills and uploads enabled, open Plugins, Skills, Create, and Upload from your computer. Choose a skill package format accepted by the file picker. You can also choose Create with editor and enter the original name, description, and instructions.
- After the scan completes, check that output-quality-rubrics appears in Installed. Type @ in a new chat, select the skill, and use the fictional example on this page to check the output.
Example task
Create a three-level rubric for fictional library-hours answers. The source says Friday closes at 18:00. A says 18:00 and cites the rule, B says Friday evening without evidence, and C says 20:00. Give anchors and example scores.What to check
Distinguish wrong, correct but vague, and correct with evidence. Give observable criteria and leave an example for independent calibration.
Original SKILL.md on GitHub · Updated
Frequently asked question
What if two reviewers score very differently?
Identify how they interpreted the same wording. Replace it with a checkable condition and rescore the disagreement example.