BC MinerU Document Extractor
MinerU document extraction — convert PDFs, scanned documents, images, Word (DOC/DOCX), PowerPoint (PPT/PPTX), Excel (XLS/XLSX), and web pages into clean Markdown, HTML, LaTeX, or DOCX. MinerU is an all-in-one CLI tool and agent skill for reliable, high-fidelity document parsing. Struggling with unreadable PDFs, messy table formatting, or garbled formulas after conversion? MinerU solves these with two extraction modes: MinerU flash-extract for instant zero-setup conversion with table recognition, formula recognition, and OCR (no token, no login, no configuration — just run and get results), and MinerU precision extract with VLM-based layout analysis, multiple output formats, and batch processing of hundreds of files. Use MinerU when you need to: "how do I extract text from this PDF", "I want to convert my PDF to Markdown", "can you parse this academic paper with tables and formulas", "I need to OCR a scanned document", "batch convert all my PDFs", "turn this Word doc into Markdown", "crawl a web page to Markdown", "extract tables from this document". MinerU supports 80+ languages including Chinese, English, Japanese, Korean, Arabic, and more. Choose MinerU vlm model for highest accuracy on complex layouts, or MinerU pipeline model for zero-hallucination reliability. Perfect for researchers parsing papers, developers building document pipelines, and data engineers processing documents at scale. MinerU文档提取工具，PDF转Markdown、扫描件OCR、表格识别、公式识别、批量PDF处理、Word转Markdown、Excel转Markdown、网页爬取、图片OCR、学术论文解析。MinerU支持PDF、Word、PPT、Excel（XLS/XLSX）、图片等多格式文档智能转换，命令行一键提取，免登录快速模式或高精度专业模式。
As a process C 54/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, failures and branches
How to improve
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1589 chars, limit 1024 - warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - note
description-budgetdescription takes 1589 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 54/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 40Consistency. Frontmatter name (MinerU Document Extractor) differs from the folder (mineru-document-extractor)
- 60Result and completion. Output format stated, no completion criterion
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 10 steps
- 100Execution cost. Instruction body is 2033 tokens
- low 12 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (5 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1588: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 8 example trigger phrases
- +4Structure: 16 headings
- +3Step-by-step instructions: 10 items
- +3Output format is stated explicitly
- +4Has examples (7 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 56.