mineru
Parse PDFs, Office files, images, and HTML into Markdown/structured outputs with MinerU. Use when SciForge or Codex needs document parsing/OCR for scientific papers, supplementary files, PDFs, scanned documents, tables, formulas, or URL/local-file parsing through MinerU standard or Agent lightweight APIs.
- License 许可
- Apache-2.0
- 所属仓库
- AGI4Sci/SciForge
- 最近更新
- 2026-11-07
- Stars
- ★ 84
- Forks
- 21
01 前言 Front Matter
- name
- mineru
- description
- Parse PDFs, Office files, images, and HTML into Markdown/structured outputs with MinerU. Use when SciForge or Codex needs document parsing/OCR for scientific papers, supplementary files, PDFs, scanned documents, tables, formulas, or URL/local-file parsing through MinerU standard or Agent lightweight APIs.
02 资源文件 3
- 参考文档agents/openai.yaml186 B
- 参考文档references/api.md2.6 KB
- 脚本scripts/mineru_parse.py11.8 KB
03 SKILL.md 正文
MinerU
Use MinerU to turn PDFs, Office files, images, and HTML into Markdown and structured outputs for downstream research workflows.
Choose API Mode
- Prefer standard mode for scientific papers and reproducible project work. It requires
MINERU_API_KEY, supports files up to 200 MB / 200 pages, batch upload,pipeline/vlm/MinerU-HTML, tables, formulas, and zip outputs. - Use agent mode for quick, small, non-sensitive parsing. It needs no token, is IP-rate-limited, supports a single file or URL, returns Markdown only, and is intended for lightweight agent workflows.
Do not store API keys in skill files, source code, logs, or committed .env files. Read standard API credentials from MINERU_API_KEY.
Quick Commands
Use the bundled CLI:
python .codex/skills/mineru/scripts/mineru_parse.py standard-file path/to/paper.pdf --output-dir outputs/results/mineru/paper --model-version vlm --extra-format html
python .codex/skills/mineru/scripts/mineru_parse.py standard-url https://example.com/paper.pdf --output-dir outputs/results/mineru/paper --model-version vlm
python .codex/skills/mineru/scripts/mineru_parse.py agent-file path/to/small.pdf --output-dir /tmp/mineru-small
python .codex/skills/mineru/scripts/mineru_parse.py agent-url https://example.com/small.pdf --output-dir /tmp/mineru-small
For local files in standard mode, the CLI requests a signed upload URL, uploads the file, polls the batch result, downloads the result zip, and extracts it.
For standard URL mode, the CLI submits the URL task, polls it, downloads the result zip, and extracts it.
For agent mode, the CLI downloads the returned Markdown file.
Recommended Defaults
- Use
--model-version vlmfor scientific papers with formulas, tables, and figures. - Use
--model-version MinerU-HTMLonly for HTML source files. - Use
--pagesto limit expensive or slow parses during tests. - Keep raw PDFs and downloaded zips in ignored local folders; commit only extracted, reviewed Markdown/metadata when appropriate.
Resources
scripts/mineru_parse.py: deterministic CLI wrapper for MinerU API calls.references/api.md: concise API reference extracted from the local MinerU HTML documentation.
04 同仓库更多 Skills
openspec-propose
Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
openspec-explore
Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
openspec-archive-change
Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.
openspec-apply-change
Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
research-memory
Draft, confirm, and synchronize research status, evidence, blockers, and review-ready GitHub summaries using the Research Memory skill, GitHub conventions, and file templates.
openspec-propose
Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.