学术论文搜索：在 Agent 里完成检索、引用与 OA 全文整理
介绍 SkillHub 上的 @baitongai/academic-article-search（学术论文搜索，v1.0.0）。该 Skill 由百通AI服务平台发布，无需 API Key，支持 Semantic Scholar、OpenAlex、Crossref、百度学术多源回退，以及 APA/MLA/Chicago/GB7714/BibTeX/RIS/Harvard/EndnoteXML 八种引用；默认输出可点击链接的 HTML，并生成 xlsx。文中说明安装方式（SkillHub CLI 或按 skillhub.md 指引）、典型对话触发词与 OA 全文、知网相关使用注意。
Read Moredsh-paper-digest：按质量信号自动生成中英文论文 Markdown 日报
dsh-paper-digest 是 juzikexue 维护的 DSH 插件，在设定时间自动检索中英文论文，用期刊档次、内容证据、时效等可审计信号排序，生成主题分组 Markdown 日报。支持 CNKI 期刊 RSS、OpenAlex、arXiv 等多源，可选本机模型中文速览与 OA PDF 下载，并在工作区创建可追问的日报会话。配置位于设置→论文日报；需 clone 后 pnpm build:client，再以 link 或 dsh plugin add 装入 web profile。MIT 0.1.1。
Read More论文搜索专业版：七源文献检索与可调深度
介绍 SkillHub 社区 Skill「论文搜索专业版」（@user_3c6cb52e/paper-search-pro-m7pa，v1.0.0）：在 OpenAlex、Semantic Scholar、CrossRef、PubMed、arXiv 与 NSSD、yiigle 等最多七个来源检索论文，支持 Quick scan 至 Audit prep 的可调深度，覆盖中文核心/CSSCI 与国内期刊及中科院分区、JCR Q1 等筛选语境。说明 skillhub CLI 安装、--dir 指向各 Agent skills 目录的方式，并给出与官方简介一致的自然语言用法示例；requires_api_key=false。
Read Morearxiv-paper-searcher：在 Agent 里完成 arXiv 检索与结构化分析
arxiv-paper-searcher 是 SkillHub 社区技能（v1.0.2，@user_eb2ccafa/arxiv-paper-searcher），默认串联 arXiv 搜索、摘要分析、热点统计与趋势判断，并返回结构化报告；用户确认后可基于同一检索条件创建定时监控。无需 API Key。本文介绍其定位、能力、通过 skillhub CLI 安装到 Cursor 等 skills 目录的方法，以及发起分析与开启监控的典型对话方式。
Read MorePubMed 文献检索：在 Agent 里查摘要并尝试全文
介绍 SkillHub 社区 Skill「PubMed 文献检索」（@user_72111f74/pubmed-search，v1.0.1）。面向 PubMed 30M+ 文献做摘要检索，并支持 auto full-text；标签 requires_api_key=false。说明 skillhub CLI 安装、Cursor/Claude Code/Codex 的 skills 目录、典型对话式用法，以及社区维护与全文获取等注意点。
Read More科研助手：134 个科学技能库覆盖的科研全流程 Skill
科研助手（@zcwl/scientific-research-assistant）是智创未来在 SkillHub 发布的企业级 Skill，v1.0.1，基于134个科学技能库，官方说明覆盖从文献检索到论文发表的全流程，核心包括文献检索与综述（PubMed/Google）。需通过 skillhub CLI 安装到各 Agent 的 skills 目录，并注意 requires_api_key。本文依据 SkillHub 页面与安装文档整理安装路径、能力边界与使用注意，不编造未公开的子功能。
Read More论文检索 · 文献速查：在 Agent 里按关键词拉取高被引论文
介绍 SkillHub 社区技能「论文检索·文献速查」（@user_13555fe3/paper-search1，v1.0.0）。输入关键词、作者或领域可查询高被引论文，返回标题、年份、引用量、DOI 与下载链接；标签 requires_api_key=false。说明 skillhub CLI 安装到 Cursor/Claude Code/Codex 等 skills 目录的步骤，以及对话中的典型请求方式与使用注意。
Read More科研技能鉴赏：科研相关 Skill 的总入口与路由
科研技能鉴赏是 SkillHub 上的科研技能总入口与路由（@user_164f4c1f/keji-skill-showcase，v1.2.0）。当你不确定该用哪个科研技能时，可先启用它：精选 13 个技能覆盖文献检索与下载、知识库、CAJ 转 PDF、引用核查、学术诚信、组会、新颖性、答辩、方法论、技能锻造与 AI 易读化等。每个技能标注适用边界、能做什么、不能做什么、该不该装，并列出触发词。安装需按 skillhub.cn/install/skillhub.md 将技能装到 Cursor 等客户端的 skills 目录，无需 API Key。
Read Moredsh-zotero：Zotero 文献全文检索插件
dsh-zotero 是一款连接 DeepSeek Harness 与 Zotero 7 的插件，旨在解决本地文献 AI 阅读难题。它通过索引 Zotero 库中的 PDF 全文，支持在侧边栏进行检索和逐块阅读，也能在对话中直接提问让 AI 查阅文献。安装需先克隆代码并执行 `npm install`，使用时需确保 Zotero 开启本地 API 并支持 OCR 以解析扫描件。
Read MoreDeepSeek Harness Plugin: ln1m/dsh-tool-literature (Literature Search Tool)
DeepSeek Harness developer Ln1m introduced the `dsh-tool-literature` plugin to address the inefficiency of agents directly executing shell commands. The plugin encapsulates the literature search workflow and exposes a structured interface for direct model invocation. Built on the `paper-search.py` script, it integrates OpenAlex citation ranking with an arXiv relevance scoring algorithm. It depends on libraries such as `@deepseek-ai/cordis` to support automated literature retrieval. Notably, the repository has been archived and merged into `dsh-tool-suite` and is retained for historical reference only.
Read Moredsh-kb-rag: Local Literature Knowledge Base
`dsh-kb-rag` is a local literature RAG plugin for DSH users, designed to address difficulties in citation tracking and source location. It indexes PDFs and Zotero libraries into a SQLite database and returns precise paragraph-level evidence with section, page number, and DOI links. The plugin supports local-first embedding and reranking, offers an MCP service mode, and is suitable for researchers who require high privacy and precise citations. Note that installation may involve large model files and limitations with scanned PDFs.
Read Morexby-article: DeepSeek Harness Literature Retrieval and Management Plugin
`xby-article` is a web-enabled plugin for DeepSeek Harness (DSH), designed to address challenges in academic research, including multi-source retrieval, cumbersome full-text acquisition, and fragmented journal evaluation. Built on the FastMCP framework, it provides integrated capabilities for literature search, full-text retrieval, reference management, literature relationship analysis, and journal quality assessment. Users can configure API keys, specify search strategies (such as comprehensive, precise, or preprint modes), and invoke the relevant APIs to quickly retrieve literature metadata and full texts, while using the built-in deduplication mechanism to intelligently analyze citation networks. Once installed, it is available for use in the DSH Web Profile.
Read MoreDeepSeek Harness Literature Search Plugin
The DeepSeek Harness (DSH) plugin `dsh-literature-search` is designed to address the lack of unified interfaces across multiple literature databases. It supports searching through PubMed and Google Scholar and normalizes the results into a standard `paper` data structure. Its features include PubMed search, single-paper details, related articles, as well as Google Scholar search and citation generation. A DSH restart is required after installation. For use, it is recommended to configure `NCBI_API_KEY` and SerpApi credentials to improve rate limits and stability.
Read MoreDeepSeek Harness Plugin: dsh-academic-paper-search
`dsh-academic-paper-search` is an academic literature search plugin designed specifically for DeepSeek Harness. It serves as a thin adapter layer that reuses the `nature-academic-search` service, providing capabilities such as cross-database search, deduplication, and citation graph retrieval. The plugin requires configuration of environment variables (such as `PUBMED_EMAIL`) and pinned dependency versions to ensure persistence and reproducibility, aiming to simplify the integration of multi-source academic data in agent development.
Read Moredsh-paperlens: Literature Search Research Copilot Plugin
`dsh-paperlens` is a Cordis plugin in the DeepSeek Harness (DSH) ecosystem for scientific research. Built on the “everything is a plugin” principle, it runs entirely locally to ensure data privacy. It offers core capabilities including multi-source retrieval, citation formatting, reference database management (SQLite), and batch export in RIS/BibTeX. It supports searches via OpenAlex and CrossRef, provides multiple citation styles such as APA and GB/T 7714, and supports scoring and tag management for paper metadata. The plugin can be installed with a command without manual configuration and does not require browser authentication, making it well suited for efficient literature retrieval, citation generation, and local archiving within the DSH environment.
Read More@onpeople/dsh-academic-research: An Evidence-Driven Bilingual Academic Research Plugin for DeepSeek Harness
`@onpeople/dsh-academic-research` is a standalone plugin for DeepSeek Harness (DSH) designed to reduce hallucinations in academic research. It requires no private database connections and leverages public APIs such as Crossref and OpenAlex to provide bilingual, evidence-based literature retrieval, review, and writing assistance. The plugin includes literature discovery, record reading, and research integrity rules to ensure reliable and trustworthy outputs. It can be installed via the `dsh` command and is currently in developer preview.
Read Moredsh-arxiv-search: An arXiv Paper Search Plugin for DeepSeek Harness
DeepSeek Harness (DSH) introduces an arXiv search plugin that requires no API key or Python dependencies. The plugin registers an `arxiv_search` tool and returns structured metadata—such as paper titles, abstracts, and links—via the public API. Users can retrieve results by entering keywords, specifying a sorting method, and setting the number of results, making it suitable for developers who need real-time access to academic literature. To install it, run the `dsh plugin` command and restart the service.
Read More@amphilagus/dsh-literature: DeepSeek Harness Literature Retrieval and Targeted Tracking Plugin
`@amphilagus/dsh-literature` is a literature retrieval and tracking plugin within the DSH ecosystem. It requires the "Literature Tracking Assistant" preset and is not loaded by default in standard coding sessions. Its core functions include retrieving topics or researchers (requiring ORCID) via Crossref and arXiv; writing results to a new global SQLite database after manual grading; and providing session-level expiry reminders and continuous tracking based on a schedule. The plugin registers three runtime skills—`literature-survey`, `literature-tracking-setup`, and `literature-tracking-search`—and supports search caching, researcher profile management, and SCI journal directory queries.
Read Moredsh-zotero-wiki: Convert Zotero library to searchable Obsidian Wiki
dsh-zotero-wiki is an MIT-licensed DeepSeekHarness (DSH) plugin designed to integrate Zotero literature management, PDF parsing, full-text note generation, and Agent retrieval into a unified workflow. It automatically syncs Zotero records from local or web sources, utilizes MinerU to parse PDF original text, and invokes DeepSeek to generate structured Chinese notes, ultimately compiling them into a searchable Obsidian Wiki. Core features include dual-source synchronization, interactive range selection, parsing quota control (default 50 items per run), Wiki index compilation, LINT self-checking, and Agent queries based on in-memory full-text retrieval to ensure answers include traceable literature citations. The plugin supports CLI and DSH native tools.
Read MoreConnecting DeepSeek Harness to a traceable academic research workbench with dsh-scholar
DSH-Scholar is a DeepSeek Harness community plugin maintained by lzszq, licensed under MIT. It is categorized under the memory category on the catalog page, which emphasizes retrieving, reading and summarizing literature within the agent loop; while the repository describes it as a research workspace for pure computational research, integrating materials, conversations, experiments, evidence ledgers and TeX manuscripts into the same recoverable project. It is gate-only by default, and agents cannot approve关卡 (levels/checkpoints) on their behalf. This article checks the installation commands, standalone workbench, Runner and typical slash commands against the catalog page and GitHub repository.
Read More