DSH Memory
Cited memory over lossless session logs with human-auditable facts linked to source events.
DSH memory plugins give DeepSeek Harness persistent recall across sessions — storing notes, indexing a codebase or knowledge base, and retrieving the right context back into a conversation.
Cited memory over lossless session logs with human-auditable facts linked to source events.
Repositories with traceable evidence of a DeepSeek Harness relationship. They have not been reviewed, install tested, or security checked — the signal below each one is the whole of what the registry currently knows.
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.
A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
Shared local memory, session history search, cross-agent handoff, and uncommitted code recovery across 43 coding agents — Claude Code, Codex, Cursor, OpenCode, Gemini CLI, OpenClaw & more (CLI & MCP). One Go binary, zero LLM calls.
Composable, view-based context for DeepSeek Harness. Pluggable sources and strategies, with layered memory and self-evolving skills out of the box.
为 DeepSeek Harness 带来「跨会话长期记忆 + 后台自我进化」能力的纯插件实现:五轨记忆 · git 分支感知 · 回合内自我审查 · 技能自我进化与技能管理器 · 四轨待办 · COI 调度 · 会话广播 · 会话搜索 · 提示词管理器 · 临时信息便签——零核心修改、零运行时依赖,随装随用、卸载即净。
LLM-supervised persistent memory for AI agents — graph-based recall, cross-session knowledge, single binary. Works with DeepSeek Harness, Claude Code, OpenClaw, and any agent runtime.
Deepseek Harness、Openclaw知识图谱记忆插件。2026年4月受邀发布在清华大学讨论会。Knowledge Graph + Memory;Knowledge Graph Context Engine for OpenClaw — extracts structured triples from conversations, compresses context 75%, enables cross-session experience reuse
白箱AGI架构探索:元认知(自我认知循环)、持续学习(知识飞轮)、世界模型(条件空间+语义时空图)、自我改进(自举纪律)、零LLM白箱管线与可审计信任护栏。
Structural memory for AI coding agents. Bi-temporal graph, MCP-native, zero LLM calls. Cursor · Claude Code · Codex · DeepSeek Harness · Hermes · VS Code · Windsurf.
🐋 DSH 鱼的记忆:给会话一份会淡忘的长期记忆,每条都带时间。名字里的「鱼」来自 DeepSeek 的鲸鱼。
Cross-session memory plugin for DeepSeek Harness: seven-layer SQLite store (soul/user/project/fact/lesson/topic/rules), BM25 retrieval, per-window dream consolidation. 跨会话七层长期记忆插件。
A DeepSeek Harness (dsh) plugin — turns dsh web into a SillyTavern-style roleplay frontend: character cards, prompt presets, lorebooks, personas, BM25 long-term memory, world-state deltas, and rollback-able floor operations, all on the dsh agent runtime.
DeepSeek Harness 长篇角色扮演agent(DSH酒馆插件):SillyTavern 角色卡导入、行动选项卡、分支对话管理、长篇记忆、关键词与语义混合检索、交互式角色卡创作、世界书、多角色 Agent 集群、文风预设、小说与角色卡导出、一键安装。 / Roleplay Agent 、Tavern plugin for DSH (DeepSeek Harness): branch conversation, long-term memory, hybrid keyword and semantic retrieval, SillyTavern cards, interactive card authoring, card export, one-click installers.
Agent 通用长期记忆层:Agent 将重要信息按类型记录为多篇可读的 markdown 记忆文档,文档以链接形式列在限长索引中,索引随会话自动注入。DeepSeek Harness(dsh)有专用插件 dsh-engramory。
Proactive associative memory for DSH: system-prompt recall before the model speaks, three-layer auto-consolidation, skill crystallization, and Astra-style context management - handoff ledgers, PLAN whiteboard, water-level sensing. Local-first, model-agnostic, zero deps. 事件驱动的主动联想记忆+Astra 式上下文管理:自动唤回/自动沉淀/技能固化/交接账本与白板跨窗口续命/水位感知。
🌌 Mebular 忆云 · 智能体共享记忆与任务协作 | One shared memory and task graph for all your AI agents and devices: automatic or on-demand, secure, controllable, and decentralized!
🧠 给 AI 一个可信的第二大脑:本地 SQLite · 先审核再入库 · 不可篡改审计 · 记忆账本。🕷 蜘蛛侠皮肤 · 🔍 自建集 R@5 99.1% / LongMemEval-S 97.2% · 🧩 一个库接入 DSH / Claude Code / Codex / Cursor / 任意 MCP · ⭐ MIT 开源 · 🚀 零依赖开箱即用
DeepSeek Harness 破甲:默认提示词面向国模「鸢码」,完整开放版,无付费墙,技能库与知识库全含。破解市面上收费 388u 的版本,免费公开。Jailbreak for every model — swap prompts per model. 求 Star 收藏 ⭐
Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness: a ctx.memory capability seam with a zero-dependency SQLite provider, a memory tool, and frozen snapshot injection. Active (unfrozen 2026-10-12); see README for alternatives.
Noema long-term memory plugin for DSH: durable, inspectable agent memory with recall tools and a settings page.
DSH Web内目前最无感的消息撤回、重编辑插件,原版体验,兼容性强,功能简单可开关,设置丰富,现代化轻量ui框架。The most seamless message recall & re-edit plugin for DSH Web — native experience, strong compatibility, simple toggles, rich settings, modern lightweight UI. DSH Web で最もシームレスなメッセージ撤回・再編集プラグイン——ネイティブ体験、高い互換性、シンプルなトグル、充実した設定、モダンで軽量な UI。
Trustworthy memory for DeepSeek Harness — 4-state trust machine, evidence chains, audited injections. Zero servers, just Markdown.
DeepSeek Harness (DSH) 插件:在模型读取前撤回已发送的文字消息,支持单条与全部撤回。A DeepSeek Harness plugin that recalls sent-but-unread text messages before the model reads them — one-by-one or all at once.
DeepSeek-Harness (DSH) 插件:✨ 提示词一键增强(5 模式·模型链·记忆链)+ 💬 语音输入增强(热键·说完自动停·口水词清理·语音纠错)
Recent memories stay detailed. Older ones grow concise. Important ones live on.近期记忆保持详细,旧记忆逐渐精简,重要记忆长期存在。
Causal memory layer for AI agents — MCP server that records decision→outcome relationships. Survives compaction.
给 DeepSeek Harness 的本地记忆层:养大肥鱼的可以下载来玩,经过测试下载之后比单纯用RAG和外部记忆库更有人情味更自然了。
Knowledge base & RAG plugin for DeepSeek Harness (DSH): chunking, local embeddings, hybrid search, management panel
DeepSeek Harness (DSH) all-in-one plugin: balance whale widget, 106-clip transparent desktop pet, QQ message card via SnowLuma, Bilibili interaction, local redacted memory. 一个 DSH 插件:小鲸鱼余额挂件 + 桌宠 + QQ 消息卡 + B站 + 双向脱敏记忆;数据只存本机。
Persistent project intelligence and memory plugin for DSH: architecture analysis, cross-session context, TODOs, and optional hybrid retrieval
DSH 原生抖音助手:有限自主观看、共享人格与兴趣记忆、原生设置和观看日报。
A temporal context graph, a memory API, retrieval primitives, and a multiple-platform integration mesh — designed to be embedded into any host process.
为明日方舟的长篇剧情打造的RAG类DSH插件,拥有多种快速检索能力。
Local-first cross-session long-term memory for DeepSeek Harness — 22 narrow tools, hybrid FTS5 + local-vector retrieval, evidence gate, provenance audit, SM-2 review. Zero external services, zero API keys.
消息撤回插件:回到发送该消息时的状态 DSH Message Recall Plugin: Return to the state when the message was sent
DeepSeek Harness 增强插件:上下文压缩与 Dream 记忆、Ultra 多智能体、OpenCU 电脑操作、代码理解、扩展管理、人格和主题。 All-in-one plugin for DeepSeek Harness: context compaction and memory, multi-agent Ultra, computer use, code understanding, extensions, personalities and themes.
DeepSeek Harness (DSH) local-first cross-session project memory: zero bundled runtime deps, bounded cache-friendly Hot Memory, BM25 recall/cache, provenance, lifecycle and bilingual Web GUI; no embeddings or external service. / DSH 本地项目记忆插件。
DSH host RSI for local LLMs (currently Qwen3.8 27B): per-use continual self-improvement, counter-reason self-check, critic escalation, and memory retrieval.简单来说根据论文建立的长期反思教训机制,让大模型不再犯错,越用越强!重点是千问3.8 27B/FLASH这种本地LLM模型!当然在线大模型理论上也可以通用!
星汐:把整理好的记忆档案连成一张看得见的星图
万象知识库:DeepSeek Harness 本地知识库插件,支持一键会话存入知识库,让你的deepseek与你一起无限进步! 好用的话求个星星!
融合 Claude Code 的 Auto Memory 与 Codex 的 Session 记忆整理,为 DeepSeek Harness 提供简单、透明、上下文友好的长期记忆。Claude Code-inspired Auto Memory meets Codex-inspired Session consolidation, bringing simple, transparent, and context-efficient long-term memory to DeepSeek Harness.
将千星沙箱(原神千星奇域)知识库接入 Deepseek Harness 的插件
Read-time project memory plugin for DeepSeek Harness (dsh)
Cross-session memory for DeepSeek Harness — plain markdown files, no database, no worker.
让大肥鱼在 DSH 中长期生活:在B站社交、生活(未来扩展多平台),具有长期记忆、可塑造人格、自然好友关系
Mnemosyne:证据、时间、情境、状态演化、情绪与纠错演化记忆系统
Local document knowledge base for DeepSeek Harness: library_add/remove/list, hybrid semantic+keyword library_search with diversity re-ranking, relevance filtering and lost-in-the-middle avoidance, citation-aware injection, library_cite_check and library_diagnose — SQLite-backed index via the storage domain, local embedding, zero model downloads.
鱼 · 单角色人格与按人记忆:可纠正、有印象、能导入旧聊天记录,日程自走与主动开口;直连 SnowLuma 收发 QQ(群默认拒绝,要列白名单或开「所有群」才回;没发出去的回复排队重发),B站 / 抖音小号私信与自己刷视频;按功能选模型并统计用量花费,能听会说 QQ 语音、收藏表情包、学群里的梗,每周记忆存档与可选周记,可选上网查。DSH 插件(单角色 feiyu)
把 WorkBuddy 的文件式记忆系统移植进 [DeepSeek Harness](https://www.deepseek.com/harness/) —— 为 Harness 提供**跨会话持久化、人类可直接编辑的 Markdown 记忆**。
Long-term memory plugin for DeepSeek Harness (dsh): four-channel retrieval fusion and continuous memory evolution. Memory construction and retrieval are based on MemGAS (ICLR 2026).
给 DeepSeek Harness 的长期记忆插件:每轮聊完自动沉淀,数据存在本机 SQLite
Installable DeepSeek Harness knowledge base plugin
让 DeepSeek Harness 拥有跨会话长期记忆:AI 自动记住你是谁、你的项目和偏好,新会话直接带上背景,生活与工作记忆自动分开互不干扰,零配置无感运行 | Long-term memory for DeepSeek Harness: the AI remembers who you are, your projects and preferences across sessions, keeps personal and work memories separate, zero-config.
Unified context for DeepSeek Harness (dsh): migrate pi/opencode/codex/claude-code/WorkBuddy sessions, share AGENTS.md, skills, project & agent memory through symlinks, and migrate automations. 统一上下文,一份上下文给所有 Harness。
面向 DeepSeek Harness 的 OpenViking 检索、资源管理、自动召回与会话记忆插件
DSH 科研插件集:论文精读与追问、Idea 知识库、SSH GPU 监控,以及按工作区管理的研究目标、假设、DAG 和实验台账。
神经元记忆网络 让 DSH 记住「为什么这么决定」,而不只是结论。思维链记忆插件:8 个池分级遗忘、每轮注入、神经元 3D 星云浏览、夜巡自动整理。
Tree-shaped knowledge bases for DeepSeek Harness: a right-side canvas you edit by hand, plus tools the model reads and writes. Nodes carry a trigger line saying when they apply; a base can be injected in full or as an index.
本地优先的文献知识库 RAG(DeepSeek Harness 插件):混合检索正文与图注,把模糊记忆定位到具体段落与图表,DOI 一键直达原文。Local-first literature RAG for DSH — turn a fuzzy memory into an exact passage/figure, one-click DOI to source.