Memory and knowledge · DeepSeek Harness plugins — Page 6

DeepSeek Harness plugin that ships the Code2KB Agent Toolkit: extract evidence-checked business and engineering knowledge from pinned source revisions, then validate and publish a searchable knowledge base.

ZCode-style workspace memory for DeepSeek Harness: one central SQLite store partitioned by workspace, index injection, and sole-writer gated writes.

Find the doc your agent just wrote. Lists the DSH workspace's Markdown, images and video in the sidebar, ranked by relevance to the current conversation, with inline preview — and hands the same ranking to the agent as a tool, returning the passage that matched in each document. Zero model calls, zero network.

分层 Token 优化管道 v2:输出阶梯 / 结果缓存 / 文件 diff / 工具裁剪 + MCP 懒加载 / 压缩调度 / 缓存命中率,基于 DSH 真实插件 API

Cross-session memory for DeepSeek Harness — plain markdown files, no database, no worker.

webui：视图图块/消息导航 + 工具调用聚合 + Markdown 渲染 + 模型推理等级同步 + AnySearch 网页搜索（合并 dsh-tool-summary / dsh-better-markdown / dsh-web-search-anysearch）

EvoResearch 的 deepseek-harness 插件实现：科研项目工作区、科研记忆、多智能体团队、AutoSkills、定时任务、多通道、WebUI 扩展与 i18n（TypeScript / Node.js，不使用 Python 与 deepagents）

Mebular memory, task graph, and shared Console for DeepSeek Harness

DSH-native HarmonyOS dev assistant: hdc device debugging, an input-row development panel, official-first versioned API docs, DevEco compile assistance, and standalone Command Line Tools support.

Terminal-style input history for the DeepSeek Harness web composer: edge-first arrows with exact draft/caret restore, browser-local persisted history, Ctrl+R reverse search, workspace-scoped recall, and fully configurable keys — plus sliding-context awareness (compaction summaries join recall/search, a compaction notice with one-click /compact fill) layered on the ordinary composer draft only.

L0~L3 分层蒸馏记忆插件 for DeepSeek Harness：自动捕获对话（L0）、抽取原子记忆（L1）、整合场景块（L2）、蒸馏核心画像/团队方法论（L3），并在模型步骤前自动召回注入。移植自 MemoryCore (TencentDB Agent Memory) 的管线设计。

祖宗记忆库记忆插件: 自研 SQLite 引擎 + MCP stdio 桥 + 自动记忆钩子 + 记忆库管理卡片 UI。零外部 Python 依赖, 可插拔。

OpenViking retrieval, resource management, auto-recall and session memory for DeepSeek Harness.

Knowledge-base stage view plugin for dsh-trading (issue #34 / P5): registers the 中栏「知识库」tab via the tradingStageViews service — Obsidian-style force-directed knowledge graph with search/tag/author/credibility filters and a card detail drawer. Graph logic comes from the @dsh-trading/knowledge pure library; bridge access goes through the shell's tradingBridge service (ctx.inject only, no cross-package module imports).

Knowledge base data plane for dsh-trading: KnowledgeCard data model, validation, graph builder, atomic file store, and knowledge_ingest / knowledge_search agent tools.

Obsidian Memory plugin for DeepSeek Harness — persistent memory via local Markdown vault

Local-first memory & context harness for coding agents

Execution-discipline plugin suite for DeepSeek Harness: anti-stuck retry interception, forced reflection, todo discipline, memory dedup, and self-evolving skills. Zero dependencies, never touches DSH source.

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

SGME 仓库级 DSH 插件包装（真实插件包在 adapters/dsh/sgme-bridge/，dsh-sgme 已发布 npm）。本文件让 dsh plugin add github:freehul/sgme 直接可装，并让 dshfind 能推导出 npm 安装命令。

Persistent project memory for dsh agents: index docs (PDF/Markdown/text) and code symbols into a searchable per-workspace store, recall them with cited sources, and keep experience entries (problems -> solutions) searchable on demand.

本地记忆（Local Memory）· Memory panel for the DSH web settings page: read, search, append and carefully edit a file-based memory archive.

记忆核心（Memory Eternal）：自研的 DeepSeek Harness 记忆插件，不移植任何既有记忆框架——对话结束后自动沉淀知识卡到本地 Markdown Vault（自研去重、自研 CJK 检索、可 git 管理），设置页提供图形化知识库（统计 / 搜索 / 知识图谱 + 侧边栏一键弹窗），Agent 通过 memory_recall 工具按需召回历史上下文。零人工干预。

Co-Engram team memory plugin for DeepSeek Harness (dsh): native Cordis tools + dynamic prompt-signals injection.