Extract text, tables, images, and OCR output from PDF documents.
npx skills add https://github.com/234194027-cpu/xianclaw --skill pdf-234194027-cpu
Please help me install this Agent Skill. Skill: pdf Source: https://github.com/234194027-cpu/xianclaw/tree/main/resources/openclaw/config/skills/pdf Command: npx skills add https://github.com/234194027-cpu/xianclaw --skill pdf-234194027-cpu
This skill removes the friction of working with PDFs by enabling fast extraction, structured table/text retrieval, and automated form annotation or filling instead of manual copy-paste and retyping.
Use case example: You receive a scanned contract bundle and need searchable text plus key fields placed onto each page; use this skill to OCR the pages, extract needed content, and generate an output PDF with the requested text positioned correctly.
Ask the skill to extract all text and tables from document.pdf and return the results page by page.
High-intent search queries and answers about installing and using this skill.
To extract text and tables from a PDF, this skill uses layout-aware parsing libraries like pdfplumber and pypdf to retrieve structured content page by page. It detects tabular structures and returns searchable text for downstream analysis or reporting.
Yes, you can fill non-fillable PDF forms by using coordinate mapping from the PDF structure or visual estimation. The skill adds text annotations at precise locations, enabling automated form filling for both fillable and non-fillable inputs.
To make a scanned PDF searchable, this skill applies OCR using pytesseract and pdf2image to process document images. It converts scanned content into searchable text, enabling document digitization and full-text search.
Available PDF transformations include merging, splitting, rotating pages, watermarking, and password protection using tools like qpdf and pdftk. These utilities support batch processing workflows for managing multiple PDF documents efficiently.
Yes, the skill supports optional font setup specifically for CJK rendering in generated PDFs using reportlab. This ensures that Chinese, Japanese, and Korean characters are correctly displayed when creating or annotating output documents.
The best way to extract images and metadata from a PDF is by using deterministic parsing libraries that read the document's internal structure. This skill retrieves embedded images and document metadata without manual copy-paste or retyping.
rvdbreemenExtracts and manipulates text, tables, images, and forms in PDF documents.
alibabaExtract text, tables, and metadata from PDF documents into JSON.
ZJU-REALAnalyze and transform raw datasets into statistical insights and business reports.
legoutExtracts text and structured data from images using OCR engines.
