yisyeasy-crypto/academic-writing-dna-skill — explained in plain English
Analysis updated 2026-05-18
Collect papers from a target journal and generate a style DNA file so your AI writes drafts that match that journal's tone.
Build a personal library of writing style profiles for different co-authors and switch between them when drafting.
Teach students academic writing by extracting style patterns from exemplary papers and showing what makes them effective.
Distill your own past papers into a DNA file so an AI assistant can draft new sections in your voice.
| yisyeasy-crypto/academic-writing-dna-skill | agricidaniel/gogh | betta-tech/harness-sdd | |
|---|---|---|---|
| Stars | 46 | 46 | 46 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | easy |
| Complexity | 2/5 | 3/5 | 2/5 |
| Audience | researcher | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Installation is just linking or copying the skill folder into the right directory for Claude Code or Codex, Python is optional and only adds precision.
``json { "i": 0, "repo": "yisyeasy-crypto/academic-writing-dna-skill", "stars": 46, "lang": "Python", "ok": true, "char_count": 2600, "text": "This project lets an AI assistant learn a specific author's academic writing style and then produce new text in that same style. You collect a handful of papers from a target author or journal, drop them into a folder, and ask the AI to study them. The AI reads through the files, produces a profile called Academic-Writing-DNA.md that captures how that author writes, and saves it. After that, when you ask the AI to write something, it checks for an existing DNA file and offers to use that style.\n\nThe skill is not a single file but a bundle of several coordinated pieces. A main entry file tells the AI what steps to follow. A knowledge folder holds reference documents: a blacklist of overused academic phrases, structural templates for different paper types, a quality checklist, and guidance on reading quantitative measurements. A script written in Python measures concrete details like sentence length and citation density. An examples folder shows sample inputs and outputs. The README stresses that copying only the main entry file will reduce the skill to a plain prompt, so the full folder structure matters.\n\nInstallation works with two AI coding assistants: Claude Code and Codex. For Claude Code, you link the entire skill folder into a skills directory and the assistant picks it up automatically. For Codex, you copy the full folder into a specific location. The README provides terminal commands for Linux, macOS, and Windows, and explains how to verify that installation succeeded.\n\nTo use it, you place one to ten papers in formats like PDF, docx, markdown, or text into a folder, then tell the AI to distill the writing style. The AI reads everything, optionally runs the Python script for precise numbers, consults its reference documents, and generates the DNA file. It then reports how many papers it processed and runs an eight point quality self check. When you later ask it to write about any research topic, it detects the saved style and asks whether to apply it. You simply agree, and it follows the rules stored in the DNA file without needing further instruction about paper type, target journal, or word count.\n\nIf Python is not available, the skill still functions but with lower precision, relying on the AI's own estimates rather than measured numbers. The README notes that the tool is intended for learning writing styles, building personal writing resources, and teaching, and should not be used to impersonate authors, plagiarize, or evade plagiarism checks.", "error": null } ``
An AI skill that studies a set of academic papers, extracts the author's writing style into a reusable DNA file, and then writes new text matching that style. Works with Claude Code and Codex assistants.
Mainly Python. The stack also includes Python, Claude Code, Codex.
The license for this repository is not specified in the available information.
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.