Repolode
heise3

heise3/academic-deai

A model-agnostic Agent Skill for Chinese and English academic editing that removes formulaic framing while preserving evidence, citations, and author voice.

Stars
158
Forks
8
Open issues
0
Last commit
12 days ago

What is it?

Academic DeAI is an Agent Skill for editing academic writing in Chinese and English. It targets stock phrases, empty openings, repetitive closing lines, and mechanical paragraph structure, and it asks the editing model to check facts, numbers, citations, terminology, and the strength of evidence before changing a sentence.

The skill is a set of text instructions and reference files rather than a tied-to-vendor tool, so the chosen model carries out the edits. It is designed to be copied into the skill folder of Codex, Claude Code, Cursor, Gemini CLI, or OpenCode. Optional Python scripts check structural differences and content locks between an original and a revised text; they require Python 3.10+.

The repository includes synthetic bilingual examples showing before-and-after revisions, along with a validation document that states what the checks do and do not establish. Tasks covered include proofreading, language polishing, academic translation, and structural revision within the scope the user specifies.

Key features

  • 01Removes formulaic openings, repetitive closings, and mechanical paragraph patterns
  • 02Keeps numbers, confidence intervals, citations, and terminology intact
  • 03Works with Chinese and English manuscripts
  • 04Installs as a folder in several Agent platforms' skill directories
  • 05Includes optional Python checks for structural and content differences

Use cases

  • Polishing a Chinese research paper's discussion section without changing its findings
  • Editing an English abstract to reduce generic framing while keeping analytic limits
  • Translating a manuscript into English while preserving citations and section structure
  • Preparing responses to reviewers with clearer, more direct wording

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/heise3/academic-deai.git
  2. 2

    Create the Codex skills directory

    mkdir -p ~/.agents/skills
  3. 3

    Copy the skill folder into place

    cp -R academic-deai ~/.agents/skills/academic-deai

Things to know

  • It is a text-rule skill run by the chosen model, not standalone software, so results depend on the model used.
  • Optional validation tools require Python 3.10+, and DOCX extraction cannot keep all styles, comments, or tracked changes.
  • The included examples are synthetic demos without independent blind evaluation.

Topics

FAQ

What is academic-deai?
A model-agnostic Agent Skill for Chinese and English academic editing that removes formulaic framing while preserving evidence, citations, and author voice. Academic DeAI is an Agent Skill for editing academic writing in Chinese and English. It targets stock phrases, empty openings, repetitive closing lines, and mechanical paragraph structure, and it asks the editing model to check facts, numbers, citations, terminology, and the strength of evidence before changing a sentence.
Is academic-deai free to use? What license does it have?
academic-deai is open source under the MIT license. Check the license terms before using it in commercial products.
Is academic-deai still maintained?
Actively maintained. The last commit to academic-deai was on Oct 8, 2026.
What can I use academic-deai for?
Polishing a Chinese research paper's discussion section without changing its findings Editing an English abstract to reduce generic framing while keeping analytic limits Translating a manuscript into English while preserving citations and section structure Preparing responses to reviewers with clearer, more direct wording
How do I install academic-deai?
Clone the repository: git clone https://github.com/heise3/academic-deai.git Create the Codex skills directory: mkdir -p ~/.agents/skills Copy the skill folder into place: cp -R academic-deai ~/.agents/skills/academic-deai
Where can I get academic-deai?
The source code and installation instructions are on GitHub: https://github.com/heise3/academic-deai

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects