Extract structured markdown text and per-page PNG images from PDF files.
npx skills add https://github.com/KaihuaTang/Index.skill --skill extract-pdf
Please help me install this Agent Skill. Skill: extract-pdf Source: https://github.com/KaihuaTang/Index.skill/tree/main/.claude/skills/extract-pdf Command: npx skills add https://github.com/KaihuaTang/Index.skill --skill extract-pdf
It converts PDFs into Claude-readable content without losing structure, figures, tables, or scanned-page visuals that naive text extraction often mangles.
Run the extract-pdf skill on your PDF at a given path or URL, writing outputs into an extraction folder, for example by instructing Claude to execute the provided command on your target PDF input.
High-intent search queries and answers about installing and using this skill.
Extracting structured markdown from a PDF uses pymupdf4llm to preserve headings, lists, and tables, outputting page-aligned markdown files. This ensures the original document layout is accurately represented for downstream summarization and analysis.
Claude can read charts and equations from scanned PDF pages because this skill renders each page into a PNG image using PyMuPDF. This enables multimodal visual interpretation of complex figures, math, and scanned content alongside the extracted text.
Parsing complex PDF tables without losing formatting is best achieved through a dual-extraction approach. This skill uses pymupdf4llm for structured markdown capture and pdfminer.six as a complementary plain-text fallback to recover text from tricky layouts or misread typography.
This PDF extraction method works with both locally saved files and URLs. You can instruct the skill to execute its rendering command on a target PDF input path or URL, which generates an extraction workspace containing a manifest.json plus aligned markdown and PNG files.
PDF text extraction mangles encoding and tricky layouts when relying on a single parser. This skill mitigates the issue by applying pdfminer.six as a complementary plain-text extractor to recover text when the primary markdown extraction misreads typography or complex document structures.
AlvaroBianoExtract text from PDFs and scanned documents into Markdown or structured data.
quanpersie2001Process multimodal media to extract text, transcriptions, and structured data.
adiytharpansaExtracts text from images, screenshots, and scanned documents using OCR.
heysuhasExtract text, tables, and images from scanned PDFs and office files.
