grading-guide-render

Render a byte-stable grading guide from validated questions.json and rubrics.json.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/AKCqhzdy/dse-subject-grading --skill grading-guide-render
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grading-guide-render
Source: https://github.com/AKCqhzdy/dse-subject-grading/tree/main/skills-v3/grading-guide-render
Command: npx skills add https://github.com/AKCqhzdy/dse-subject-grading --skill grading-guide-render

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Deterministically render a grading guide from validated questions.json and rubrics.json into a byte-stable rubric/{year}/grading_guide.md. Pure template — no LLM calls. The guide serves as the standard reference for Phase 5 sub-agents and must be byte-stable across runs for identical inputs.

Core Features & Use Cases

  • Deterministic rendering of grading guides from validated inputs (questions.json and rubrics.json).
  • Byte-stable output with atomic writes and no timestamps for reproducibility.
  • Optional incorporation of level descriptors and subject notes when present.
  • Template-driven per-question rendering supporting nested structures and images.
  • Use case: Generate a consistent grading guide for a given subject and year to onboard automated grading workflows.

Quick Start

Run the render script with the questions.json and rubrics.json to generate rubric/{year}/grading_guide.md.

Frequently Asked Questions about grading-guide-render

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

FAQPage Schema
How do I generate a deterministic grading guide from JSON rubric data?

To generate a deterministic grading guide, run a template-render script with validated questions.json and rubrics.json inputs to produce a byte-stable markdown document without any LLM calls.

What makes a rendered grading guide byte-stable across multiple runs?

A rendered grading guide achieves byte-stability by using deterministic templating, omitting timestamps, and applying atomic writes, ensuring identical JSON inputs always produce identical output files.

Can I include level descriptors and subject notes when rendering an educational assessment?

Yes, the rendering process supports the optional incorporation of level descriptors and subject notes into the grading guide when these fields are present in the validated JSON inputs.

Does the grading guide template-render process require any LLM calls?

No, the grading guide template-render process is purely template-driven and requires no LLM calls, relying entirely on deterministic templating to produce the standard reference document.

What is the best way to onboard automated grading workflows for a specific year and subject?

The best way to onboard automated grading workflows is to generate a consistent grading guide from validated JSON inputs, serving as a byte-stable standard reference for automated grading sub-agents.

Are there limitations to rendering grading guides with nested structures and images?

The template-driven rendering supports per-question rendering with nested structures and images, but is limited to validated JSON formats and deterministic output constraints, requiring strict input validation.