DeepSeek插件商店0研究与知识中的文档处理插件。
This plugin integrates Univer office capabilities into DeepSeek Harness so an agent can create, edit, import, and export spreadsheets, documents, presentations, and related content from natural language requests with in-conversation preview and review.
This is an office content plugin for DeepSeek Harness that lets users create, edit, and export spreadsheets, documents, and presentations, using built-in `univer_*` tools with in-conversation preview and approval flows.
通过配置的 MinerU 服务将 PDF、图片、DOCX、PPTX 和 XLSX 等本地文档解析为结构化 Markdown 或 JSON。
Provides session-isolated file uploads and a read_document tool for helping models read text, PDF, DOCX, and XLSX documents.
A local document-auditing plugin that checks academic manuscripts for AI-like style patterns, evidence integrity, and target-journal writing characteristics during editing.
选择并整理已加载的对话内容，导出为可下载或复制的PNG图片和Markdown文件。
A document-processing tool for creating, parsing, checking, formatting, revising, and auditing Chinese official DOCX files under GB/T 9704 rules.
Provides agent-oriented file uploads, attachment references, document-to-Markdown conversion, image OCR, and paged document reading.
为 better-sidebar 注册 DOCX、XLSX 和 PPTX 文件预览器，在浏览器中提供 Office 文档预览。
Provides DeepSeek Harness tools to read, create, edit, audit, merge, and split spreadsheets, PDFs, presentations, and Word documents.
Parses local PDF, Office, Markdown, HTML, CSV, and text files and performs offline OCR on images and scanned documents.
This plugin is a novel-writing assistant for DeepSeek Harness that manages local chapters and offers text analysis, style and continuity checks, plot and setting tracking, and writing-assistance features.
Provides agent-facing tools to create, read, and update Word, Excel, and PowerPoint files within the session workspace.
A text-processing plugin that reads bundled writing guidance before generating or polishing prose and checks preservation of content anchors.
Converts semantic JSON presentation definitions into editable PPTX files with validation, themes, previews, and agent-oriented CLI workflows.
将 Word、PPT、Excel、OpenDocument、RTF、EPUB、CSV 和 PDF 等文件转换为 GitHub-Flavored Markdown。
将课件和电子书内容整理为可打印的思维导图HTML，并可生成配套交互式测试题。
通过可配置的 PaddleOCR 服务提取图片和 PDF 文本，并解析文档布局、表格及结构。
将 DSH 文档整理为支持搜索、翻译、目录导航和对话面板的本地网页阅读界面。
为智能体提供对电子表格、Jupyter 笔记本、PDF、DOCX 和 PPTX 文件的受限读取，以及对 XLSX 和 IPYNB 文件的创建与编辑。
Parses PDFs, Office documents, HTML, and images through MinerU APIs and saves structured Markdown results to the workspace.
A DeepSeek Harness plugin that adds parsing, OCR, format conversion, image rendering, and long-document indexing for PDFs, Office files, text, and other attachments.
PaperLab is an AI-assisted LaTeX paper revision workbench that applies PDF annotations, validates compilation, and manages document versions with Git.
将 EPUB 或 PDF 书籍解析、阅读并转换为可安装的 SKILL.md 技能文件，流程包含 OCR、智能体辅助分析和人工审批。
提供 HTML 与 Markdown 双向转换、GFM 表格规范化以及 Markdown 目录生成功能。
通过 Harness 文件系统本地读取 PDF，提取文本、标题元数据、页数和指定页码范围。
将 DeepSeek Harness 智能体会话自动导出为 Markdown 文件，便于复盘、写作和审计。
Processes localization files by inspecting, extracting, protecting tags, slicing text, and losslessly reassembling game scripts and common documents.
将 DeepSeek Harness 当前会话导出为可移植的 Markdown 和结构化 JSON 文件。
Extracts, normalizes, and checks DOI, arXiv, and web citations in drafts while producing bounded evidence results.
Uses local Tesseract OCR to convert attached images into text before model requests, with an optional mode for passing original images through to vision models.
通过 PaddleOCR-VL 解析 PDF 和图片的文档布局，支持异步任务进度跟踪，并将结果导出为 Markdown 和图片文件。
A DSH plugin for generating, visually editing, compiling, and exporting PPTX presentations.
将当前 DeepSeek Harness 对话导出为 Markdown、PDF 或长 PNG 图片。
Analyzes Markdown documents by generating heading outlines and reporting hierarchy skips, duplicate headings, H1 issues, and unclosed code fences.
Generate self-contained interactive HTML reports from Markdown, tables, charts, China province maps, flowcharts, math, and drill-down tables — a DeepSeek Harness (dsh) plugin
Renders Markdown as standalone HTML and provides a callable tool plus a web drawer for previewing, editing, and exporting documents.
Parses PDFs, images, Word documents, and PowerPoint files into structure-preserving Markdown and JSON through the SoMark API.
根据教师提供的教材、课程标准和Office模板，生成同步的英语教学PPTX与DOCX教案。
用于基于已有理论分析和数值验证完善学术论文，并辅助处理引用内容。
Review Markdown documents in a local browser with comments, replacement suggestions, and deletion suggestions that the agent can read and process.
Preview and edit Markdown files in a sidebar, add block-level or text-range annotations, and format them as structured revision instructions for the chat input.
用于通过百度云或本地兼容 OpenAI 的服务识别图片及多种文档，并将长文档解析为 Markdown。
Recognizes text in local images through configured OpenAI- or Anthropic-compatible vision APIs and returns confidence scores with bounding-box coordinates.
面向法律事务的工作区插件，提供文件索引与搜索、文档预览和提取、DOCX/PDF 生成与修订、可选 OCR 及任务跟踪功能。
将 DeepSeek Harness 会话中的任务、统计、工具调用和产出整理为可保存、可验证并可选发布的日报、周报、交接文档或文章草稿。
Transforms selected DSH conversations into structured reports with covers, summaries, and chapters, exportable as PNG, HTML, PDF, or Markdown.
Uses a local office engine to read, create, and edit PPTX, DOCX, XLSX, and PDF documents.
This plugin parses Word, PDF, or plain-text question files, extracts questions and images, and generates printable exams from selected items with blank answer space, outputting both editable DOCX and printable HTML views.
Preview produced Markdown, image, code, and text files directly in DeepSeek Harness conversations with fullscreen viewing, content copying, and system-app opening.
Automatically aligns Markdown tables in assistant replies after each turn and provides a command to clear the model-visible conversation history.
This is a DeepSeek Harness web plugin that lets users attach files to a session and use built-in tools to inspect, read, OCR, and render uploaded content.
This plugin adds a `read_pdf` tool to DeepSeek Harness so AI models can read PDF files in-session and receive extracted plain text for summarization or analysis.
This is a DeepSeek Harness plugin for bid rejection checks that lets users upload tender and bid documents, automatically extracts rejection items, runs typo and logic-risk checks, and outputs reports in PDF, HTML, or Markdown.
This plugin lets you open and edit .xlsx files in the DSH right sidebar, with cell editing, live formula updates, formatting controls, and saving changes back to the original file.
在当前工作区中解析 PDF、Office/OpenDocument 文件和图片，并将结果输出为 Markdown、文本或 JSON。
This plugin provides DeepSeek Harness agents with tools to create, read, query, and edit .docx/.xlsx/.pptx files, plus in-web preview support through the dsh sidebar.
Translates text and files with configurable languages, writing styles, and glossaries while preserving formats and optionally writing results back in place.
为 DeepSeek Harness 提供支持多语言、术语表、格式保留、文件导入和本地历史记录的可配置文本翻译工作台。
用于在本地解析 PDF、Office、文本和图像文件，并通过离线 OCR 提取扫描文档内容。
Queries Crossref for DOI metadata, formats references in several citation styles, generates BibTeX entries, and checks DOIs found in text.
Converts prompts, text, or Markdown into themed HTML presentations, editable PPTX files, and JSON manifests.
将 Claude Code 的技能、规则和指令文件预览并导入 DeepSeek Harness，支持用户级或项目级范围及冲突处理。
为纯文本模型提供图片理解、OCR 及 PDF、Office、表格和 HTML 文档解析，并将结果转换为可供模型使用的文本或 Markdown。
通过多轮交互澄清需求，并生成包含流程、权限、校验、状态和接口约定的可交付交互文档。
Adds DeepSeek Harness attachments with drag-and-drop uploads, file cards, and text extraction for PDF, DOCX, XLSX, TXT, and related files.
将主题、文档或大纲转换为经过内容规划、视觉审核和质量检查的可编辑多页 PowerPoint 演示文稿。
将参考图片重建为包含可编辑文本、形状、线条和独立图片资产的单页 PowerPoint，并生成预览与质量检查报告。
A DSH web plugin that extracts text from uploaded or dragged PDF, DOCX, PPTX, plain-text, and code files into the chat composer.
一个为 DeepSeek Harness 提供的文档生成插件，可根据项目结构、代码签名和 Git 历史生成 README、API 文档、变更日志及 JSDoc 注释。
A DeepSeek Harness plugin for reading local PDFs, extracting document metadata and text, and rendering pages as PNG images.
Provides academic writing support through paper outlines, title and abstract skeletons, citation formatting, phrasing checks, submission checklists, and common LaTeX snippets.
A translation workflow toolkit that provides text segmentation, terminology and glossary handling, source-target quality checks, tone guidance, and workspace-scoped translation memory.
A DSH web GUI sidebar plugin for browsing, rendering, editing, and saving local Markdown files.
Analyzes local images and documents through CLI or HTTP model services and can generate image files from text prompts.
Converts DeepSeek Harness sessions into Markdown or self-contained HTML documents for local preview, copying, downloading, and sharing.
将 DeepSeek Harness 会话转换为可预览、复制或下载的 Markdown 或自包含 HTML 文档，并提供可选工具调用信息与启发式脱敏。
Provides MCP-based policy consultation, authoritative search, deep research, and formal document drafting and delivery.
This is a DeepSeek Harness web plugin that enhances chat attachment handling by accepting additional document and image formats and injecting processed content into the composer flow as document-style attachment handling.
This plugin for DeepSeek Harness provides workbook inspection, image rendering, diffing, and editing for `.xlsx`/`.xlsm` files, with an editing flow that includes validation and confirmation steps.
A Chinese writing assistant plugin that provides cn_polish text formatting and cn_stats metrics for word count, structure, and reading time in chat workflows.
This plugin adds attachment support to the DeepSeek Harness chat UI, allowing users to upload or drag and drop files and images and display them as file cards, while passing extracted text into the message context when available.
A document-processing plugin for DeepSeek Harness that reads PDF text page by page and applies OCR to scanned or image-heavy pages.
An agent artifact tool for creating, managing, and previewing self-contained HTML documents in an in-app tab.
Use SendPage MCP to publish, update, and manage shareable HTML documents, with export support for PNG, PDF, and DOCX.
This is a DeepSeek Harness conversation plugin that provides a visual slide editor mode with AI annotation subagents, enabling multi-slide deck editing, workspace-backed loading/saving, and presentation-style export from the active session.
Converts Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF files into line-numbered Markdown that can be read in pages.
一个基于纯提示词的文档处理技能包，可根据代码库和 Git 历史生成 README、PR 描述、changelog，并提供结构化代码审查意见。
Uploads or references local text, code, Word, Excel, and PDF files, extracts their text, and attaches it to the next chat message.
An offline macOS OCR plugin that uses Apple Vision to extract text from screenshots, scans, and document images for model use.
dsh-slides is a DeepSeek Harness plugin tool that converts structured talk outlines into self-contained HTML slide decks and can also export an editable PPTX version.
An on-demand DeepSeek Harness plugin that recognizes text in local images using macOS Vision or Windows Media OCR.
This plugin adds an Export Chat action in DeepSeek Harness Web to produce a session-only Markdown export for reading and handoff.
This DeepSeek Harness plugin reads PDF papers in a side reader panel, converts them into selectable structured text with figure and equation rendering, and supports asking questions on selected text.
将主题、文档、表格、图像或现有演示文稿转换为可编辑的 PowerPoint 幻灯片，并提供主题选择、本地预览和导出流程。
Extracts text locally from PDFs, Office documents, images, and scans with offline OCR, full-text search, and batch processing.
This plugin adds a local PaddleOCR tool to DeepSeek Harness, enabling the `/ocr` command to process workspace PDF/image files and output extracted results as Markdown files under artifacts/paddle-ocr.
This plugin adds a `read_pdf` tool to DeepSeek Harness that reads a workspace PDF and extracts its text page by page for agent usage.
This is a Chinese contract risk review skill that scans contract content for risk clauses and returns a graded risk report with excerpts and suggested fixes.
This plugin generates structured meeting minutes from transcript text or audio input, extracting discussion points, decisions, and action items with assignees and due dates.
This plugin provides Markdown table-of-contents generation in dsh, helping structure and navigate Markdown documents more easily.
Scans local project metadata, directory structure, scripts, technology indicators, and license information to generate a structured README.md.
A DSH plugin for creating and managing novel projects, chapters, and world settings through chat and a web interface, with data stored locally as JSON.
面向 Markdown 知识库提供本地优先的 SEO、GEO 和 AEO 审计、内容规划、来源检查及受版本保护的内容更新。
将技术提案通过检索、专利要点提炼、起草、审阅、制图和多格式导出流程整理为专利披露文档的 AI 插件。
Provides hosted MCP tools for compressing and optimizing PDF files and inspecting page count, file size, and password protection status.
Generate daily, weekly, monthly, or custom-range reports from local DeepSeek Harness sessions, save them as Markdown, and optionally sync them to flomo.
A read-only tool that checks HTTP(S) links in Markdown documents and reports dead links with their file locations.