grading-answer-extraction

Extracts per-student answer text from PDF files into structured JSON using extraction-v2.

Updated Aug 27, 2026

Search Tags:#workflow #pdf #json #grading #data-processing #extract

Ready to run

npx skills add https://github.com/AKCqhzdy/dse-subject-grading --skill grading-answer-extraction

💬 Prompt for ChatGPT / Claude Chat▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the extraction of per-student answer text from PDF submissions to streamline grading and ensure consistent RawItem formats.

Core Features & Use Cases

Page-by-page extraction: invokes extraction-v2 one page at a time to preserve handwritten or semi-structured responses.
Structured outputs: writes per-student JSON to extracted/{year}/students/student{N}.json (RawItem answer schema) and optionally saves page images when flagged.
Resume-safe and validation: skips processing when outputs are newer than the source, and supports recovery after interruptions.

Quick Start

Invoke the extraction workflow to generate per-student answer JSONs from data/masked_data/${SCOPE_SUBJECT}/${SCOPE_GRADING_YEAR}/student_answers/ and store results under extracted/{year}.

Frequently Asked Questions about grading-answer-extraction

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

FAQPage Schema

How do I extract per-student answers from PDF files for grading?▼

To extract per-student answers from PDF files for grading, this Skill processes student{N}.pdf submissions page by page, preserving handwritten or semi-structured responses to generate structured per-student JSON outputs.

Does the extraction process support resuming after an interruption?▼

Yes, the extraction process is resume-safe; it automatically skips already processed PDFs when the existing output JSON files are newer than the source documents, allowing recovery after interruptions.

What is the page-by-page extraction mode used for PDF answer processing?▼

Page-by-page extraction mode invokes extraction-v2 on one PDF page at a time, ensuring that handwritten or semi-structured student responses are accurately preserved in the resulting JSON files.

Can I save page images alongside the extracted student answer JSONs?▼

Yes, when needed, the workflow can optionally save page images as PNG files alongside the per-student answer JSONs, storing them in the extracted directory structure for reference.

How do I organize input PDF submissions for the grading answer extraction workflow?▼

You should organize input PDF submissions as student{N}.pdf files placed in the data/masked_data/${SCOPE_SUBJECT}/${SCOPE_GRADING_YEAR}/student_answers/ directory for the workflow to process them correctly.

What format does the extraction workflow output for each student's answers?▼

The extraction workflow outputs per-student answer text in a structured JSON format following the RawItem schema, writing each student's data to extracted/${SCOPE_GRADING_YEAR}/students/student{N}.json.