PyMuPDF with JS

PyMuPDF compiled to WebAssembly - Full PDF manipulation in the browser

Extract text from PDFs/scans (pymupdf, marker-pdf).

MCP server that renders PDF pages as images (JPEG/PNG) using PyMuPDF

DeepSeek Harness plugin: content-aware PDF reading tools for vision models (pdf_scan / pdf_read_page / pdf_render_region). Detects figures (vector+raster), tables, math, and two-column layout per page, then routes text pages to extraction and figure/table

英文文献精读插件：注册 paper_ingest 工具（按 URL 抓取论文网页文本），并内置 paper-reader skill（PDF 提取 / 句子切分 / 全文逐段翻译、解析、逻辑链条梳理与方法论/亮点提炼工作流，含环境自检）。

The fastest Node.js PDF library — 0.8ms mean, 5× faster than the industry leaders, 100% pass rate on 3,830 real-world PDFs. Prebuilt native bindings, no build toolchain: text extraction, Markdown/HTML conversion, PDF creation and editing.

The fastest WebAssembly PDF library — 0.8ms mean, 5× faster than the industry leaders, 100% pass rate on 3,830 real-world PDFs. Zero-dependency for Node.js, browsers, and edge runtimes: text extraction, Markdown/HTML conversion, search, form filling, crea

DSH plugin: translate text-based PDFs via DeepSeek or any OpenAI-compatible API, preserving layout, fonts, images and links (editable output PDF, ≤50 pages per task).

Elenchus & Crucible — Socratic problem discovery & architectural stress-testing engine for OpenCode and Oh My OpenAgent. Multi-agent research with Kitchenham-light SLRs, Search Inversion Matrix, ATAM quality scenarios, and 6-vector stress spikes.

This package wraps the `pdf2docx` Python package with `node-calls-python` for interoperability with Node.js.

[FIPS 140-3 validated build] The fastest Node.js PDF library — 0.8ms mean, 5× faster than the industry leaders, 100% pass rate on 3,830 real-world PDFs. Prebuilt native bindings, no build toolchain: text extraction, Markdown/HTML conversion, PDF creation

PostgreSQL-based memory search engine for AI agents

One-command installer for dsh-kb-rag (local literature knowledge-base RAG for DSH): `npx dsh-kb-rag-install` does everything — Python deps, engine smoke test, pnpm, DSH plugin install+activate, model pre-download (auto HF mirror retry).

基于 MinerU 的论文阅读 MCP Server(文/图/表 + 出 pre)。Python 实现,通过 uv 启动;npx 即用。

Agent skill for intelligent content extraction: YouTube/Bilibili subtitles, web articles, local files (PDF/Word/EPUB/text), and audio/video transcription via whisper.cpp. Extraction only - no LLM calls.

MCP server for permanent PDF redaction. Exposes redact_pdf and check_pdf; rasterises and rebuilds each page so redacted text is gone, not covered. Never uploads the document.

Model-facing compare_docs tool for DeepSeek Harness: line-level diff between two documents (txt/md/docx/pdf) with unified-diff rendering

OCR extension powered by PP-OCRv6 — extract Chinese and English text from images using Python PaddleOCR.

MCP server for HTML/Markdown/PDF conversion, local OCR, PDF classification, PPT generation, and document-to-Markdown conversion