플러그인
DeepSeek-Harness 플러그인을 찾아보고, 필터링하고, 설치하세요.
플러그인 311개 찾음
dsh-markdown-reader
s1lencewill/dsh-markdown-reader
작업공간 Markdown 파일을 GFM, KaTeX, Mermaid, 목차 탐색, 파일 실시간 다시 로드, 브라우저 인쇄/PDF 내보내기를 지원하는 독립형 전체 화면 리더로 렌더링합니다.
dsh-table-attach
yybukn/dsh-table-attach
붙여넣기 또는 드래그 앤 드롭으로 표와 문서(xlsx, csv, pdf, zip 등)를 DSH 작성창에 첨부합니다. 파일은 세션 작업공간에 저장되고 `@` 참조가 삽입됩니다.
dsh-file-explorer-kit
ice5kysl/dsh-file-explorer-kit
세션 내 “Files” 탭(`conversation.view`, 순서 20)에서 활성 세션의 작업공간 디렉터리를 이동 경로와 함께 탐색하고 Markdown(`marked` + `DOMPurify`), 이미지, 줄 번호 텍스트, PDF를 인라인 미리보기합니다. `ctx.webServer`에 등록된 읽기 전용 호스트 경로 `/dsh-files`로 제공하며 쓰기 엔드포인트는 없습니다. zh/en 지원.
docindex
johnxu22786/docindex
workspace 문서(MD/PDF/DOCX/TXT)의 로컬 의미 검색 인덱스입니다. FTS5 BM25와 로컬 임베딩 하이브리드 검색, 줄 번호가 포함된 인용, 증분 업데이트를 제공합니다.
DragView
anzhaohao/dragview
파일 첨부를 Codex 스타일 카드로 드래그 앤 드롭하고, 앱 내에서 PDF/텍스트/비디오/오디오 미리보기와 기타 형식에 대한 시스템 기본 앱 열기를 지원합니다.
dsh-shareone-plugin
sudoprivacy/dsh-shareone-plugin
TXT, Markdown, HTML, PDF, Word, PowerPoint 파일을 ShareOne 링크로 게시합니다. 비밀번호, 워터마크, 텍스트 페이지 업데이트, TXT/Markdown/HTML 댓글을 지원합니다.
pdf-background-gray-codex-skill
zjsthmjialin/pdf-background-gray-codex-skill
DSH에 remove-pdf-background-gray 스킬을 등록: 이미지 기반 PDF의 회색/미색 스캔 배경을 해상도, 페이지 기하, 안티에일리어싱된 텍스트 가장자리를 유지한 채 흰색으로 만듭니다(무손실 Flate 되쓰기). 단일 Python 스크립트(pypdf + Pillow + numpy).
inspiration-deck-workshop
zjsthmjialin/inspiration-deck-workshop
DSH에 Inspiration Deck Workshop 스킬을 등록: 로컬 정적 HTML 프레젠테이션 덱(템플릿 6종, 레이아웃 25개 이상, 테마 23종과 모션 쇼룸), 검증 + PNG/PDF 내보내기 CLI, 스모크 테스트, 런타임 의존성 없음.
dsh-netcafe
mario03690/dsh-netcafe
MCP를 통한 호스팅 도구 서버: 인보이스와 은행 명세서 추출(코드 내 산술 검증 포함), 정수 센트 장부 대사, PDF와 표 변환, SQL 방언 변환, 중국 이동 네트워크 관점에서의 중국 본토 도달성 프로브.
dsh-cloudflare-browser-run
realalexandreai/dsh-cloudflare-browser-run
DeepSeek Harness 플러그인: Cloudflare Browser Run 기반 웹 브라우징 도구(markdown / 스크린샷 / pdf) — JS 렌더링, 로그인 세션, WebMCP를 지원하는 실제 헤드리스 Chrome
dsh-doc
polunzh/dsh-doc
打印工坊:把文本/Markdown 排成 A4 讲义 PDF,探测局域网打印机能力后直接双面打印,并回报出纸回执。Print Studio: typeset text into a print-ready A4 PDF and print it duplex on a LAN printer with capability-aware options.
dsh-tools-deepseek-media
oh-wang/dsh-tools-deepseek-media
Agent tools for DeepSeek OCR, ASR and TTS: recognize text in PDF and image files, transcribe WAV audio, and synthesize speech to a local file.
dsh-client-deliverable-pipeline
liuhange789/dsh-client-deliverable-pipeline
Markdown in, client-grade printable HTML report out: one self-contained file with a branded cover page, a table of contents auto-built from H2/H3 with anchors that resolve, body typography, and a running footer. Print rules are the point — A4 `@page` with 2cm margins, table rows that never split across pages, repeating table headers, headings kept with their content. Brand colour is WCAG-checked: links and TOC numbers are darkened to reach 4.5:1 on white, cover text switches between black and white by luminance, and `#rrggbbaa` is composited first. Open the file and press Ctrl+P to export the PDF; no external CSS/JS/fonts, no browser engine.
dsh-ppt-master
dashan-0079/dsh-ppt-master
Ships the PPT Master skill (upstream v6.8.0, MIT) byte for byte as a DeepSeek Harness bundle and registers it with ctx.skills as ppt-master. Adds a cross-platform CLI (doctor, install-deps, run) that handles Python discovery, UTF-8 stdio, a writable work copy and an isolated venv, so the skill turns PDF, DOCX, web pages or Markdown into natively editable .pptx decks.
dsh-paper-workbench
qianxi-00/dsh-paper-workbench
Read research PDFs with bilingual translation, native DSH conversations, and Obsidian notes.
dsh-page-images (dsh-slide-pages)
linyonghang/dsh-page-images
Renders a .pptx/.ppt/.odp attachment into one PNG per slide by exporting the deck to PDF first, posts the slides with the deck text outline, and exposes a slide_pages tool.
dsh-page-images (dsh-pdf-pages)
linyonghang/dsh-page-images
Renders a PDF attached to a message into one PNG per page and posts the pages back as a conversation message; a pdf_pages tool renders explicit page ranges at a chosen DPI.
dsh-fast-pdf
stewart222b/dsh-fast-pdf
PDF reader for the right sidebar: continuous scroll, search, outline, and selection translation.
dsh-plugin-litsearch
peijiping/dsh-plugin-litsearch
Search ten free literature sources from DSH: the model runs and refines the search while the sidebar shows results, rounds and full-text availability, and you can download PDFs, export citations or save papers to Zotero from either side.
dsh-gemini-antigravity-auth
battiatus/dsh-gemini-antigravity-auth
Gemini models from your Google Antigravity subscription inside DeepSeek Harness: Google OAuth, multi-account quotas, Google Search grounding, and image, audio, video and PDF understanding.
dsh-static-kb
charlie-yangjiaqi/dsh-static-kb
DeepSeek Harness 静态知识库:自定义分区、导入 PDF/网页/TXT/电子书并由模型转化为可检索 Markdown,按对话绑定、资料库优先回答。
dsh-plugin-paper-reader
tulipe1735/dsh-plugin-paper-reader
Bilingual reading library for papers, tech blogs and long reads: paste a link, read the English original with a Simplified-Chinese translation in one shared scrollbar, keep headings, lists, code blocks and inline images, filter the shelf by paper/blog/article, edit card metadata, and ask a DSH-model chat rail that sees the text. Translation is cached per block; a static demo-snapshot gallery can be attached to scroll-driven pages, and the Original button opens the live page. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-plugin-knowledge-base
fsnh/dsh-plugin-knowledge-base
Local knowledge base for DeepSeek Harness: a folder tree, in-app viewing of the original layout of Office files, PDFs and images, Chinese-friendly full-text search with optional vector fusion, image descriptions produced by a vision model so pictures become searchable, and three read-only tools the model calls to list, search and read files. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
DeepSeekHarness-plugin (dsh-pdf-tools)
alotherss/deepseekharness-plugin
Page-scoped PDF tools for the model: pdf_info reads the page count and whether a text layer exists, pdf_outline and pdf_search locate pages, pdf_text extracts only those pages, and pdf_render rasterises one page. Per-call caps and CJK radical normalisation are built in.