skills.rest
TZUKWAN
TZUKWANCOMMUNITY
Explore all skills by this author →
GitHub

emarx

Generates Chinese humanities academic papers with citation audits and Word delivery.

2|1|Updated Jun 24, 2026
Search Tags:#literature review#academic writing#docx generation#citation audit#cnki#chinese papers#marxist theory
One-click install
npx skills add https://github.com/TZUKWAN/emarx-skill --skill emarx-tzukwan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: emarx
Source: https://github.com/TZUKWAN/emarx-skill
Command: npx skills add https://github.com/TZUKWAN/emarx-skill --skill emarx-tzukwan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python-docx, pdfplumber, pywin32, playwright, requests, beautifulsoup4, lxml, pillow, opencv-python, numpy, ddddocr, playwright-stealth, and includes scripts (resource) and references (resource) components.

What problem does it solve? Writing Chinese humanities and social sciences papers (especially Marxist theory, ideological education, and Party history) requires evidence-grounded argumentation, strict citation discipline, and journal-specific Word formatting that generic AI writing cannot reliably produce. ## Core Features & Use Cases - Evidence-first paper production: Scans the user's workspace, anchors 3-5 real papers for close reading, builds concept ledgers and paragraph-level outlines before writing a 10000-12000 character paper. - Built-in CNKI integration: Diagnoses literature gaps and automatically searches CNKI for core/CSSCI journal papers when local sources are insufficient, with source-claim binding before citation. - Multi-agent review and audits: Runs 15 subagent reviewers at outline, section-card, and paragraph stages, plus scripts for citation audits, AI-trace audits, and DOCX audits. - Word delivery: Generates DOCX files with circled footnote markers, per-page renumbering, GB/T 7714 references, and hanging indents. - Use Case: A graduate student needs a journal-style paper on generative AI and cultural communication; the skill reads their workspace sources, supplements via CNKI, writes section by section with quality gates, and delivers an audited Word document. ## Quick Start Ask the AI to use the emarx skill to write a complete academic paper on your topic based on the materials in your workspace, then deliver it as a Word document.

Frequently Asked Questions about emarx

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

FAQPage Schema
How do I write a Chinese academic paper with AI using real sources?▼

Use this skill to scan your workspace for real papers, anchor 3-5 relevant sources for close reading, then generate the paper section by section with argument cards and quality gates. It forbids unsourced claims like generic literature review phrases.

How to generate a Word document with Chinese footnote citations?▼

Run scripts/emarx_build_docx.py on the Markdown draft to produce a DOCX with circled footnote markers, per-page renumbering, Song typeface body text, and hanging-indent references. Pandoc must be installed separately on the system.

Does the skill support CNKI literature search?▼

Yes, it includes a built-in CNKI module using Playwright that searches core and CSSCI journals when workspace literature is insufficient. CNKI papers must pass source-claim binding and original-text verification before entering the paper.

What Python dependencies does the paper generation require?▼

It requires python-docx, pdfplumber, playwright, ddddocr, and related packages listed in requirements.txt. Scripts auto-create a virtual environment and install dependencies on first run, but pandoc must be installed manually for Word output.

Why does the first script run take several minutes?▼

The first run automatically creates an isolated virtual environment, installs Python dependencies, and downloads the Playwright Chromium browser. This one-time setup typically takes 3-10 minutes and requires network access.

Related Skills

Explore More Education & Research Skills→
jrpeasejrpease
77

figma-environment-setup

Configures the local working folder and connects Claude to Figma for design-system workflows.

Community
Advanced
icerain-cmdicerain-cmd
15

responding-to-peer-review

Organize reviewer comments into traceable revision records and response drafts.

Community
Advanced
bordenetbordenet
8

wiki-debunker

Validate wiki page claims against git history, issue trackers, and meeting transcripts.

Community
Advanced
TRY-0508TRY-0508

chinese-academic-writing

Formats Chinese academic papers in LaTeX and Word with citation and typography rules.

Community
Intermediate