glmv-doc-based-writing

Convert PDF or DOCX documents into Markdown articles from user requirements.

458|40|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/zai-org/GLM-skills --skill glmv-doc-based-writing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: glmv-doc-based-writing
Source: https://github.com/zai-org/GLM-skills/tree/main/skills/glmv-doc-based-writing
Command: npx skills add https://github.com/zai-org/GLM-skills --skill glmv-doc-based-writing

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, PyMuPDF, and includes scripts (resource) components.

What problem does it solve?

This Skill automates the process of reading one or multiple PDF/DOCX documents and producing a coherent, requirement-driven Markdown text (paper, article, report, review, brief, proposal, etc.), removing the need for manual summarization and drafting from source files.

Core Features & Use Cases

  • Document ingestion: Accepts up to 50 remote PDF/DOCX URLs or local PDF files converted page-by-page into images for GLM-V processing.
  • Flexible output: Generates full Markdown content tailored to user requirements, including academic papers, articles, summaries, reviews, and proposals.
  • CLI and integration: Provides a Python CLI script with options for custom system prompts, temperature, max tokens, and saving results as .md or .json; requires a ZHIPU_API_KEY and optionally PyMuPDF for local PDF support.
  • Use case: Convert a set of research PDFs into a single structured article that highlights methodology, key findings, and implications.

Quick Start

Draft a 500-word technical article in Markdown from the provided PDF URL focusing on methodology, results, and a concise conclusion.

Frequently Asked Questions about glmv-doc-based-writing

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

FAQPage Schema
How do I convert multiple PDF documents into a single Markdown article?

To convert multiple PDF documents into a single Markdown article, you can provide up to 50 remote PDF URLs or local files. The tool processes these sources and generates coherent Markdown content tailored to your specific requirements.

Can I generate a research paper from DOCX files automatically?

Yes, you can generate a research paper from DOCX files automatically by supplying remote document URLs. The system reads the source files and drafts structured Markdown text focusing on methodology, findings, and conclusions.

Do I need a ZHIPU_API_KEY to process local PDF files into Markdown?

Yes, you need a ZHIPU_API_KEY environment variable to process local PDF files into Markdown. Local PDFs are converted page-by-page into images for GLM-V processing, which also optionally requires the PyMuPDF library.

What is the best way to draft a technical report from a batch of academic PDFs?

The best way to draft a technical report from academic PDFs is to use a CLI script that ingests up to 50 documents. It outputs full Markdown content, allowing you to set custom prompts, temperature, and max tokens for tailored results.

Does this document conversion approach support saving outputs as JSON?

Yes, this document conversion approach supports saving outputs as JSON. The Python CLI script provides options to save the generated Markdown article results locally as either .md or .json files.

What are the limitations when processing local PDF files for article generation?

When processing local PDF files for article generation, you are limited to 50 files per run and require PyMuPDF installed. The system converts local PDFs to images sequentially, which may impact processing speed for large documents.