yerayl/science-research-writing-skills — explained in plain English
Analysis updated 2026-05-18
Paste a rough draft paragraph and get it revised with notes on what changed.
Turn raw bullet points or data into a structured, drafted paper section.
Get a diagnostic audit of a section with problems ranked by severity.
Feed in a published article section to extract a reusable writing model.
| yerayl/science-research-writing-skills | akii-technologies-ltd/akii-seo-ai-search-optimizer | apex-quant-systems/polymarket-weather-trading-bot | |
|---|---|---|---|
| Stars | 50 | 50 | 50 |
| Language | — | Markdown | TypeScript |
| Setup difficulty | easy | easy | hard |
| Complexity | 1/5 | 2/5 | 4/5 |
| Audience | researcher | writer | developer |
Figures from each repo's GitHub metadata at analysis time.
This repository is a writing assistant skill designed to help researchers and graduate students produce better scientific papers. It plugs into Claude Code or the CodeX AI coding environment as a slash-command skill, meaning once installed you can invoke it from within those tools to get help drafting, revising, or reviewing any section of a research article. The skill covers all the standard sections of a STEMM paper: abstract, introduction, methods, results, discussion, conclusion, and title. The skill is built on the frameworks from Hilary Glasman-Deal's book on science research writing, which itself analyzed thousands of published articles in science, technology, engineering, mathematics, and medicine to identify the patterns that appear in successful papers. Those patterns have been turned into interactive guidance rather than a static reference. When you paste a draft paragraph, the skill returns an improved version with notes on what changed. When you give it bullet points or raw data, it drafts a structured section for you. You can also ask for a diagnostic review, which produces an audit report listing specific problems ranked by severity. Beyond draft and revise, there are two other modes. The reverse-engineer mode takes a published article section you paste in and extracts the writing model behind it, which you can then apply to your own work. The interactive guide mode walks you through a section step by step, asking questions and building the text with you rather than handing you a finished draft to accept or reject. The underlying frameworks address common weaknesses in academic writing: matching verb strength to how certain your evidence actually is, using tense consistently to signal whether you are describing your own work or the existing literature, linking sentences so the logic is clear to a reader who is not already an expert in your field, and making sure every data point comes with a sentence that explains what it means rather than leaving the number to speak for itself. Installation is a folder copy to the Claude Code or CodeX skills directory, followed by a restart. There are no external dependencies and no API keys required beyond your existing Claude or CodeX setup.
A Claude Code and CodeX skill that coaches researchers through drafting, revising, and reviewing STEMM research papers using frameworks from a published writing guide.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.