liltloom
★ 4liltloom语织：中文优先、用户可控的 AI 写作风格记忆层 / Chinese-first style memory for AI
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Adkid-Zephyr/LiltloomDiscover The Best DSH plugins
Memory and knowledge · DeepSeek Harness plugins — Page 15
2467results
Find plugins for conversation memory, knowledge retrieval and context management.
lume-dsh-plugin微光 (Lume) — DSH Desktop 人设与记忆插件：从聊天记录、小说、剧本或设定文档蒸馏角色，保留长期与临时记忆、风格学习、认可语料、人设切换纠偏、记忆星图和角色卡导入导出；任务侧仅保留一段轻量工作纪律。
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:cayan0x/Lume@123caiji/dsh-graph-memoryCross-session knowledge graph memory for DeepSeek Harness — typed nodes, causal relations, PageRank, communities, and dual-path recall (vector + FTS5).
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:123caiji/dsh-memory-toolkit#cb99d5053555f6643e1118841c24a87914e70c3c&path:packages/graph-memory@123caiji/dsh-lean-contextToken-saving orchestration for DeepSeek Harness: five orthogonally-layered capabilities (discipline, convergence, routing, audit, rebuild), each independently toggled.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:123caiji/dsh-memory-toolkit#cb99d5053555f6643e1118841c24a87914e70c3c&path:packages/lean-context@akslcw/dsh-negative-ledgerAgent negative-knowledge ledger for DeepSeek Harness: records failed paths with evidence hashes and retry conditions, auto-invalidates on evidence change.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:akslcw/dsh-negative-ledger@bingo_touth/memory-mcp-server分层长期记忆 MCP 服务器：本地 MiniLM 或 OpenAI 兼容嵌入 API，供 LLM Agent（DeepSeek Harness / Claude Code）做可语义检索的长期记忆
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:BingoAgentTouch/Personal_MCP@deepseek-ai/dsh-client-ui-memoryMemory viewer: a '记忆' settings section that renders the AI's text, KV, and mind-map memories served by dsh-tool-memory over /api/memory
@deepseek-ai/dsh-client-ui-memoryMemory viewer: a '记忆' settings section that renders the AI's text, KV, and mind-map memories served by dsh-tool-memory over /api/memory
@deepseek-ai/dsh-tool-memoryLong-term memory for DeepSeek Harness: remember/recall/list/forget tools plus a memory skill, persisted under $DSH_HOME/memory/memory.jsonl
@detpecca/dsh-llm-wikiDeepSeek Harness plugin — manage an LLM-Wiki knowledge base from the agent: wiki_search / wiki_read / wiki_stats / wiki_validate / wiki_fix / wiki_errorbook / wiki_ingest
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:detpecca/dsh-llm-wiki@dsh-external/dsh-kb-sieve知识库筛子：把文档（md/txt/docx/pdf）构建成可审计知识包（references + SQLite FTS5），提供确定性检索（kb_query）与原文精读（kb_read）工具，可生成零脚本的 DSH skill 并动态加载。
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:omdsh-dev/dsh-kb-sieve@duke-dsh-plugins/dsh-memory-managerPersistent project memory for DeepSeek Harness, ported from ZCode: ZCode-style markdown memory store (per project + global) with a full memory-management page in Settings, index-first prompt injection, agent memory tools (memory_save / memory_get / memory_list), plus one-click import from an existing ZCode memory directory.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:MoonlitDropOfBlood/dsh-memory-manager@eqman00003/knowlp-ragKnowLP-RAG: dual knowledge-graph retrieval for Markdown notes — DeepSeek Harness (dsh) bundle: MCP server + native Cordis plugin
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:wly8691-jpg/knowlp-rag@goodddgrades/dsh-behumanDurable cross-session memory and self-evolving skills for the DeepSeek Harness
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add @goodddgrades/dsh-behuman@graycode/dsh-pluginGrayCode host plugin for DeepSeek Harness: workflows, permanent memory, and workspace checkpoints
@graycode/dsh-pluginGrayCode host plugin for DeepSeek Harness: workflows, permanent memory, and workspace checkpoints
@snow-the/dsh-session-handoffSession handoff & context management for DeepSeek Harness: structured handoff documents (export/resume/status), active context pruning (acp_* tools + pressure banner), session management (trash/restore/purge), model routes for one model across vendors/plans, and optional OpenViking memory / archify diagram integration. Core is zero-dependency (official compaction API + node builtins); memory/diagram enhance when detected.
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:snow-The/dsh-session-handoff@stardustlc/dsh-suiteDSH 组合包：一次安装 18 个独立插件，覆盖邮件、日历、记忆、媒体与开发工具；更新由用户选择。欢迎提交 issues 和 PR。
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:STARDUSTLC666/dsh-suite@trisoul/dsh-apiTriSoul 中枢 API：模型配置（统一/精细两模式，settings 命名空间）+ 后台 AI 监控统计 + /trisoul/api/* 路由
@trisoul/dsh-canvasTriSoul 画布编排器（M5 + C）：agent/pre-step 步边界检查表面上下文膨胀，圈定可手术区间交给手术刀（ctx.compaction.compactRegion）；会话内状态区（恒真区只增不改 / 状态区原地改写，LLM 小作业提炼 + ctx.inject 注入，ctx.bail('trisoul/canvas-state')）；手术后探针题验收（trisoul/canvas phase=probe）
@trisoul/dsh-client-memory-uiTriSoul 记忆中枢面板（会话视图标签页）
@trisoul/dsh-client-monitorTriSoul 后台 AI 监控面板（会话视图标签页）
@trisoul/dsh-client-settingsTriSoul 模型配置（统一/精细两模式）：设置页分区 + 会话头部模式切换
@trisoul/dsh-consensusTriSoul 多体共识插件：拦截 dsh 主循环的 llm/stream，N 路盲写→匿名两票制表决（候选+融合）→胜者整份放行 / 融合胜出则全员见稿重写再表决
Find the right plugin, then get started
Search by task
Use a feature keyword or category, then sort by recent updates or GitHub stars.
Explore capabilities
Read the description, AI analysis and original README to understand what a plugin does.
Check and install
Review compatibility, dependency ranges and instructions, then copy the available install command.
Questions about DSH Hub
What is DSH Hub?
DSH Hub is a community plugin directory for DeepSeek Harness, bringing together descriptions, source links and installation instructions to help you discover and compare extensions.
Does inclusion mean a plugin is tested and compatible?
No. Detail pages show recorded compatibility and dependency ranges. Unknown does not mean compatible; check the project documentation before installing.
How do I install a plugin?
Open its detail page and check the installation section. Copy the command into your Harness environment when available, or follow the project README.