eval-grading-quality-audit

Audit per-question grading quality by running sub-agents that cross-check evidence against rubrics.

Updated Aug 27, 2026

Search Tags:#education #audit #rubric #quality-assurance #grading #evaluator

Ready to run

npx skills add https://github.com/AKCqhzdy/dse-subject-grading --skill eval-grading-quality-audit

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: eval-grading-quality-audit
Source: https://github.com/AKCqhzdy/dse-subject-grading/tree/main/skills-v3/eval-grading-quality-audit
Command: npx skills add https://github.com/AKCqhzdy/dse-subject-grading --skill eval-grading-quality-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Automates a per-question quality audit of grading outputs by deploying a sub-agent per student to read the rubric, the extracted answer, and the grading result, scoring each question on five criteria, and reporting failures and root causes.

Core Features & Use Cases

Per-question audit using sub-agents that independently re-judge student work against the rubric and verify consistency with the awarded marks.
Produces grading_quality_audit.json with per-question details, including failure_stage and root_cause_type, plus a student-level summary for remediation planning.
Integrates reference rubrics and failure taxonomy to enable traceable diagnostics and audit commentary.

Quick Start

Provide the rubric, extracted student answers, and grading outputs for the target cohort to launch per-student audits and generate the aggregation file grading_quality_audit.json.

Frequently Asked Questions about eval-grading-quality-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I audit grading quality for multiple students against a rubric?▼

Audit grading quality by deploying sub-agents that independently read the rubric, extracted student answers, and grading outputs to score each question on five criteria and report root causes. This automated per-question review ensures consistency with awarded marks across a target cohort.

What is per-question grading quality assurance using LLM sub-agents?▼

Per-question grading quality assurance uses LLM sub-agents to independently re-judge student work against a reference rubric and verify consistency with awarded marks. It enforces evidence cross-check and rejects heuristic scoring to provide traceable diagnostics and audit commentary.

How do I generate a grading quality audit report with root-cause diagnosis?▼

Generate a grading quality audit report by orchestrating independent judgments per student, scoring each question, and aggregating results into grading_quality_audit.json. This file includes per-question failure_stage, root_cause_type, and a student-level summary for remediation planning.

Can I use automated grading audit for multiple students and years of data?▼

You can use this grading audit across multiple students and years by providing the rubric, extracted student answers, and grading outputs for the target cohort. The system orchestrates independent sub-agent judgments per student and aggregates the results into a consolidated JSON file.

Does the grading audit enforce evidence cross-check instead of heuristic scoring?▼

The grading audit enforces no heuristic scoring and requires evidence cross-check against the reference rubric and failure taxonomy. This ensures all per-question diagnostics and root-cause diagnoses are fully traceable and backed by verified grading evidence.

What do I need to provide to start a grading quality audit?▼

To start a grading quality audit, you need to provide the reference rubric, extracted student answers, and grading outputs for your target cohort. These inputs allow the sub-agents to launch per-student audits and generate the aggregated grading_quality_audit.json file.