proofread

Detect prose issues in metricsAI chapter notebook markdown cells.

6|3|Updated Jan 21, 2026
One-click install
npx skills add https://github.com/quarcs-lab/metricsai --skill proofread-quarcs-lab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: proofread
Source: https://github.com/quarcs-lab/metricsai/tree/main/.claude/skills/proofread
Command: npx skills add https://github.com/quarcs-lab/metricsai --skill proofread-quarcs-lab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Proofread metricsAI chapter notebooks for text quality issues to ensure professional prose and accurate content in education materials.

Core Features & Use Cases

  • Automated detection of 9+ categories of prose issues (concatenated words, missing spaces after punctuation, doubled words, broken LaTeX, unclosed formatting, broken links, emoji remnants, camelCase in prose, and long paragraphs).
  • Proofreading score and issue breakdown with safe automated fixes and backup creation.
  • JSON output for programmatic integration and batch processing across all chapters.

Quick Start

Run the proofreading workflow on a chapter notebook (for example ch08) to generate a report of text-quality issues and optionally apply fixes with --fix.

Frequently Asked Questions about proofread

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

FAQPage Schema
How do I automate proofreading markdown cells in Jupyter notebooks?

You can automate notebook prose quality checks by scanning all markdown cells in the notebooks_colab directory to detect concatenated words, missing spaces, doubled words, broken LaTeX, unclosed formatting, broken links, emoji remnants, camelCase, and long paragraphs. It generates structured JSON reports and optionally applies fixes with backup creation.

What is the best way to batch check text quality across multiple chapter notebooks?

Batch processing text quality checks across chapter notebooks is supported by scoping the analysis to the notebooks_colab directory. The tool processes all markdown cells, outputs a proofreading score with an issue breakdown, and returns JSON output for programmatic integration across all chapters.

Can I automatically fix broken LaTeX and formatting issues in notebook markdown?

Yes, automated fixes for broken LaTeX, unclosed formatting, and other text issues are available using the --fix option. The workflow creates backups before applying corrections to ensure your original notebook content remains safe while restoring professional prose quality.

Does the proofreading workflow detect concatenated words and missing spaces after punctuation?

Yes, detecting concatenated words and missing spaces after punctuation are core automated checks. The workflow scans markdown cells to identify these specific prose-quality issues alongside doubled words, broken links, emoji remnants, camelCase in prose, and long paragraphs.

How do I get a structured report of prose issues for programmatic use?

You get a structured report of prose issues via JSON output designed for programmatic integration. The proofreading score and issue breakdown are returned in JSON format, allowing downstream pipelines to consume the text-quality results from the chapter notebooks automatically.

What types of text quality issues can be caught in education and research notebooks?

Text quality issues caught in education and research notebooks include concatenated words, missing spaces after punctuation, doubled words, broken LaTeX, unclosed formatting, broken links, emoji remnants, camelCase in prose, and long paragraphs, ensuring professional prose and accurate content.