165 results
Local knowledge base for DeepSeek Harness: index markdown / text / code / web pages into a zero-dependency SQLite FTS5 store, retrieve snippets with sources, distill session experiences — tools kb_add, kb_add_url, kb_search, kb_stats, kb_list, kb_remember, kb_ingest_session, kb_experiences + interactive Web panel (kbquery fence).
A DeepSeek Harness (DSH) plugin that reads PDF files of any size (no 64KB limit): full Unicode text extraction (Chinese/English) via pdfjs-dist, with automatic OCR for scanned or image-heavy pages (Windows WinRT OCR zh-Hans-CN + en-US, or optional tesseract.js).
Local-first issue board for DeepSeek Harness: board data, model-facing tools, an agent planning loop, and a human approval queue for agent-proposed issues.
智能公文写作 DeepSeek Harness 插件——0知识库在线调用，免费10次/7天，2元/次或399元包年，GB/T 9704 38类文种。我们是一群致力于体制内智能公文写作工具垂直领域的语言学和管理学985博士团队，理想是：技术服务思想、思想服务生活。43万字知识库是你的底气，有了它，加班写材料成为过去式。 | Official Document Writing Skill for DeepSeek Harness — 38 types of Chinese official documents (请示/报告/通知/纪要 etc.), GB/T 9704 formatting, online API knowledge (0 KB bundled), free 10 calls/7 days, then ¥2/call or ¥399/year. Developed by a PhD team in linguistics & management. Write official documents 10x faster.
OI 出题工作台：知识点锁定、数据构造、本地评测，打包 Hydro/HOJ 原生题目包，发布/管理 Hydro 与 HOJ OJ，提供 Python 外部管理 UI 与系统斜杠命令。Opt-in skill-first plugin for DeepSeek Harness.
Taskboard for DeepSeek Harness — issue kanban (sidebar panel + agent tool + managed local service)
Evidence-first multi-model coding comparison workbench for DeepSeek Harness
Local project, requirement-group, and conversation workbench for DeepSeek Harness Web and Desktop
Local knowledge base for DeepSeek Harness: document CRUD, keyword search and a kb_search model tool.
Windows-grade notifications for DeepSeek Harness: system toasts, custom sounds, and a taskbar tray badge for task completion and decision prompts — a native DSH profile plugin (bundle + client + settings namespace), zero patching of built-in packages.
Wake up DeepSeek Harness: one-click desktop/taskbar/startup shortcuts that start the dsh web server (hidden) and open the GUI, a cold-start whale-girl dance overlay, and a close-button overlay with a fully-quit / close-window dialog.
Local-first RAG knowledge tools for DSH: kb_query / kb_ingest / kb_crawl / kb_related, backed by the kb-rag Python pipeline (Ollama bge-m3 + ChromaDB + SiYuan). Zero API cost, data never leaves your machine.
Windows-first vision suite for DeepSeek Harness.
A workbench-quality docking system for DeepSeek Harness Web plugins
Approval Assistant for DeepSeek Harness: approval reminders, auto-approval rules, completion reminders and question popups with per-feature channels (popup / system notification / sound / custom sound). 审批提醒、自动审批、完成提醒与提问提醒。
将 WorkBuddy（国内版/国际版）的模型接入 DeepSeek Harness — bring WorkBuddy models into DeepSeek Harness.
一个轻量的 dsh（DeepSeek Harness）插件：安装桌面双击启动器 —— macOS 上是带官方鲸鱼图标的 dsh.app，Linux 上是 .desktop 入口。零依赖，约 147 KB。（终端命令在独立的 dsh-launcher 包）
DSH web plugin: a right-side workbench panel (file tree / file viewer / git review) with a draggable-width side panel that coexists with the official tool-details panel.
DSH Web 三栏工作台：对话流只保留 AI 输出（工具调用/思考移入右侧面板），左侧工作区文件树带 git 状态徽标。Three-column workbench for the DSH web GUI: clean conversation flow, workspace file tree with git badges, and a live tool-call/think panel.
Out-of-tree dsh web UI plugin: a session-header git workbench chip opening a drawer with the file tree, per-file diff, history, compare, staging, commit, and sync (fetch/pull/push).
Hybrid memory plugin for DeepSeek Harness: L1 snapshot memory (MEMORY.md/USER.md, read on demand, zero context cost) + L2 searchable knowledge base (one-fact-per-file, SQLite FTS5) + L3 import from Hermes/Claude Code/Codex/WorkBuddy and approved Agent Hub documents. No automatic injection. Data lives on D drive, never C.
A living memory workbench for DeepSeek Harness — watch Hindsight memory grow, read and use it, in a persistent sidebar.
写作工作台：在 DeepSeek Harness 中随时查看并调用本地 .dsh-notes 笔记资料进行写作
DSH agent tools: slice a large image into labeled tiles (size or columns x rows) with an overview and a visual workbench; read_tiles selects only what the vision model needs — by region, explicit ids, the user's workbench selection, or a named target a vision model grounds on the overview.