BB mineru-ai
MinerU AI document parser — intelligent document extraction powered by AI. Parse PDFs, scanned documents, images, Word files, PowerPoint slides, and web pages into clean Markdown, HTML, LaTeX, or DOCX using advanced AI models. Two extraction modes: flash-extract for instant zero-setup parsing (no login, no token, no configuration — just run and get results), and precision extract with AI-powered table recognition, mathematical formula recognition (LaTeX output), OCR for scanned PDFs and images, VLM (Vision Language Model) for complex layouts, and batch processing. Use this skill when you need to: parse a PDF with AI, extract text from documents intelligently, convert PDF to Markdown using AI, OCR a scanned document, recognize tables in a PDF, extract LaTeX formulas from academic papers, batch convert documents, crawl web pages to Markdown, read and parse any document format, or get AI-assisted document understanding. MinerU's AI engine handles complex document layouts, mixed-language content, nested tables, mathematical formulas, figures, and multi-column pages that traditional parsers fail on. Choose vlm model for highest accuracy or pipeline model for zero-hallucination reliability. Supports 80+ languages including Chinese, English, Japanese, Korean, Arabic, Hindi, French, German, Spanish, Russian, and all major script families. Works with local files and URLs. Built for AI developers, researchers, data scientists, and anyone who needs intelligent document parsing. Works as a Claude Code skill, MCP tool, or standalone CLI. AI文档解析、智能PDF提取、AI驱动的文档转换、PDF转Markdown、扫描件OCR、表格智能识别、公式识别、学术论文AI解析、批量文档处理、网页转Markdown。MinerU AI引擎，支持复杂排版、多语言、嵌套表格、数学公式，传统解析器无法处理的文档都能轻松搞定。
As a process B 69/100 · Nearly there — weak spots: when it triggers, inputs and preconditions, running it twice
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: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1689 chars, limit 1024 - note
description-budgetdescription takes 1689 of the ~15000-char shared budget for all skills - note
frontmatter-keyunknown frontmatter key "read_when"
Process rating: all ten parameters 69/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 6 mutating operations with no state check
- 60Result and completion. Output format stated, no completion criterion
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 44 steps
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3566 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 11 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (4 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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1688: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +4Structure: 28 headings
- +3Step-by-step instructions: 44 items
- +3Output format is stated explicitly
- +4Has examples (11 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 55.