OpenDataLab's MinerU Document Explorer skill lets an agent search across a folder of Markdown, PDF, Word and PowerPoint files and then read inside a single long document by table of contents, page, keyword or meaning. It can pull out tables and figures and can feed documents into a wiki-style knowledge base. It needs the tool installed from npm, plus Python libraries for the office formats. An optional cloud API key for parsing is set by the user.
DOCXCreates, reads, and edits Word (.docx) documents and templates with formatting fidelity.Document intelligence
PDFCreates, reads, edits, and analyzes PDF files (text, tables, forms, layouts) with Claude document skills patterns.Document intelligence
Turn answers into one-page visual HTML explainersThis skill turns a complex answer into a single visual HTML page. The agent drafts the content in extended Markdown, and a bundled command-line tool named am takes care of layout, color, dark mode, and where diagram elements sit, so no HTML or SVG is written by hand. It is meant for explanations of how something works, comparisons and trade-offs, rankings, and anything with a table or several sections. The page follows the language of your question, and its settings can be changed with a config command.Document intelligenceDiagramsHTML
Document API EndpointDocuments API endpoints with clear request/response contracts and examples.Document intelligence
Read and edit Google Docs with the gws CLIThis skill reads and edits Google Docs with gws, the Google Workspace CLI. A helper adds text to the end of a doc, and the raw API covers other document edits.Document intelligenceGoogle WorkspaceDocs
Theme FactoryCreates cohesive visual themes (palettes, type, spacing) for apps and documents.Document intelligence
