Extract text/tables, create, merge, split PDFs. Fill PDF forms programmatically. Use for PDF processing, generation, form filling, document analysis, batch operations.
Résultats pour « PDF »
Skills correspondant au nom, à la description ou à l’auteur GitHub.
Catégories: Entreprise
Explorer
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Agent needs to fill in a PDF form or programmatically process, gen...
skills-pdf
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, ge...
gembox-skill
Coding assistance for [GemBox components](https://www.gemboxsoftware.com/). Use when users ask about any GemBox component or coding task that can be performed with a GemBox component. This includes GemBox.Sprea...
prd-generator
Generate comprehensive Product Requirements Documents (PRDs) for engineering teams, optimized for the Ralph autonomous coding loop. Use when user: (1) requests a PRD or product requirements document, (2) says "...
looker
Multimodal analysis agent for PDF, images, video, audio, charts, and screenshots. Use when: analyzing documents, describing images, analyzing video/audio content, explaining diagrams, or extracting data from ch...
cheat-sheets
Expert knowledge for creating effective cheat sheets with PDF export. Activate when creating/editing cheat sheet content, designing quick reference sheets, or working with PDF export functionality. (project)
search-paper
学术论文搜索、摘要查看、PDF下载、BibTeX抓取。支持 Google Scholar、arXiv、IEEE、Elsevier。
simple-report-export
Simple report generation without Google Cloud - uses Gmail App Password or local file export (PDF, Excel, Word-compatible formats)
roadmap-generator
기적지수 분석 결과를 기반으로 30일 맞춤 PDF 로드맵 생성
biz-doc-generator
Word, Excel, PPT, PDF 등 기업용 문서를 생성하고, 결과를 docs/exports에 저장하며 임시파일을 자동 정리합니다.
pdf-automation-gotenberg
GotenbergでPDF変換・帳票出力を自動化する。バックオフィス/営業/CSの帳票フローで使う。
docling
Turn ANY File (PDF, DOCX, Audio) into LLM-Knowledge in SECONDS using Docling.
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, ge...
paper-reader
AI/CS paper reader - analyzes PDF papers and extracts key points, contributions, and methods
extract-pdf-text
Extract text content from PDF files using pdfplumber. Use when asked to "extract text from PDF", "read PDF contents", or "get text from this PDF file".
design-export
Export to PDF/HTML/markdown. Use when distributing documentation, creating presentations, or archiving design docs.
receipt-parser-engineer
Create, fix, and refine deterministic receipt/invoice parsers in curlys-books, including vendor detection, OCR text extraction routing (pdfplumber for text PDFs → AWS Textract fallback; images → Textract), gold...
debug-pdf-extraction
Debug PDF text extraction issues. Use when parser is producing wrong data, missing units, or garbled text from a PDF.
parse-new-game
Parse a new GCACW game PDF to extract scenario data. Use when adding a new game to the roster generator, extracting units from PDF tables, or setting up scenario names.
book-notes
Transcribe handwritten book notes from reMarkable PDFs or paper photos. Trigger phrase "process book notes". Handles single-source book notes with optional chapter structure. Converts handwriting to markdown, a...
deep-dive-notes
Transcribe handwritten multi-source research notes from reMarkable PDFs or paper photos. Trigger phrase "process deep dive notes". Handles research compilations with multiple sources per document. Converts hand...
historian
Data Analyst. Monitors logs in real-time and generates the PQC Benchmark Book PDF.
api-reference
PDF-RAG API reference. REQUIRED after any failed curl/jq to localhost:8000 (404, null, jq error). Also use when uncertain about endpoint path or response shape.
Catégories populaires
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.