homework-grading-workflow

Extract student names from scanned homework and generate per-student PDFs.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Sallvainian/Sallvain-Marketplace --skill homework-grading-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: homework-grading-workflow
Source: https://github.com/Sallvainian/Sallvain-Marketplace/tree/main/plugins/homework-grading-workflow/skills/homework-grading-workflow
Command: npx skills add https://github.com/Sallvainian/Sallvain-Marketplace --skill homework-grading-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires PyMuPDF, PyYAML, and includes scripts (resource) components.

What problem does it solve?

Quickly convert hand-scored or scanned homework submissions into organized, per-student PDFs and keep a live completion tracker by matching names to a roster.

Core Features & Use Cases

  • Extract student names from scanned pages using vision, match to roster (with fuzzy matching when needed), split PDFs into individual student files, and update a completion spreadsheet.
  • Session resume support for large batches (99-image per session) to minimize disruptions.
  • Provides workflow guidance and status tracking to ensure reliable verification before finalizing PDFs.

Quick Start

Provide a scanned homework PDF and roster; the system will extract names, generate per-student PDFs, and update the completion tracker.

Frequently Asked Questions about homework-grading-workflow

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

FAQPage Schema
How do I split a batch of scanned homework PDFs into individual student files?

To split a batch of scanned homework PDFs, this workflow extracts student names using handwriting recognition and matches them to a roster. It then generates individual per-student PDFs and updates a YAML tracking file and completion spreadsheet.

How does fuzzy roster matching work for scanned homework pages?

Fuzzy roster matching handles variations in handwriting recognition by comparing extracted names against your provided roster. This ensures accurate identification even if scanned handwriting is unclear, correctly generating per-student PDFs.

Can I resume grading a large batch of scanned worksheets if the process is interrupted?

Yes, you can resume grading large batches of scanned worksheets. The workflow enforces a 99-image per session limit and uses a YAML tracking file to save progress, minimizing disruptions when resuming interrupted sessions.

Do I need to manually verify student names before finalizing the per-student PDFs?

Yes, manual verification is required before finalizing per-student PDFs. The workflow provides status tracking and guidance to ensure reliable name matching verification before generating the final PDFs and completion spreadsheet.

What are the limitations of using PyMuPDF and PyYAML for automated homework grading?

Limitations include a strict 99-image per session limit for batch processing. PyMuPDF handles PDF extraction while PyYAML updates the tracking file, but you must manage session boundaries and verify fuzzy matching results for large homework batches.