dsh-trilogy
Per-project memory for DeepSeek Harness. Three Markdown files per workspace — PROJECT.md, DECISIONS.md and SESSIONS.md — are auto-created on the first step of any session in that workspace, re-read from disk and re-injected only when they change (and re-injected if compaction dropped them), and written back by classification through a memory_checkpoint tool, with a bounded end-of-turn nudge when a turn did work but recorded nothing. memory_read and memory_search read the live log and the archive, the latter over a zero-dependency BM25 index. The session log is capped at 200 entries with the overflow moved to SESSIONS-archive.md — never injected, but visible, readable and searchable, and restorable one entry at a time. Settings page: a workspace list with a path filter, read-only tabs for the three memory files plus the archive, reload, export/import of one workspace's memory as a JSON bundle, a two-step clear, a panel that rewrites or removes the AGENTS.md instruction block, and a PROJECT.md staleness reminder. A composer status chip shows that workspace's sync state and creates the files on click. Zero dependencies — no vector store, no embeddings, and no model calls.
DeepSeek Harness 的项目记忆插件。每个工作区三份 Markdown 文件(PROJECT.md / DECISIONS.md / SESSIONS.md):会话首个步骤自动创建、内容变了才重新注入(被上下文压缩挤掉也会重注入),并按分类经 memory_checkpoint 工具写回,在「干了实事却没记录」时给一次收尾提醒。memory_read 与 memory_search 可读活日志与归档,后者走零依赖 BM25 索引。会话日志上限 200 条,溢出搬进 SESSIONS-archive.md —— 不自动注入,但可见、可读、可搜,并能逐条恢复。设置界面:带路径筛选的工作区列表 / 三份记忆文件与归档的只读页签 / 重新读取 / 单个工作区记忆包的导出导入(JSON)/ 两步确认的清除 / AGENTS.md 指令段的重写或移除面板 / PROJECT.md 陈旧提醒。输入框状态图标显示该工作区的同步状态,无记忆时可点击创建。零依赖——不含向量库、embedding 或模型调用。
How to install
dsh plugin add github:TodayJin/dsh-trilogy About
dsh-trilogy **给 DeepSeek Harness 的项目记忆插件** —— 为每个工作区维护三份 Markdown 记忆文件:**自动创建、自动加载、自动记录**。 记忆不该靠模型自觉。这些行为由 **host 插件**保证,而不是靠提示词提醒模型: | 能力 | 只靠提示词/技能 | 本插件 | |---|---|---| | 新建项目自动创建三个文件 | ❌ 要人工触发 | ✅ 会话启动时自动 scaffold | | 每个会话开始加载三个文件 | ❌ 靠模型记得读 | ✅ 自动注入 + 内容摘要去重 | | 进展分类记录进三个文件 | ⚠️ 靠模型自觉 | ✅ 收尾兜底提醒 + memory_checkpoint 工具 | 「会话冷启动」与「锁定决策」是原生实现,不依赖模型自觉。 --- 三个文件 全部在 <项目根>/memory/ 下。**项目根的判定规则**: **项目根 = 会话工作目录本身。** 插件完全不看版本控制 —— 你打开哪个目录,记忆就写在那个目录的 memory/ 里。工作区有没有 .git 都不影响判定。 若你想要「一个仓库一份记忆、不要每个子包各一份」,把 projectRootStrategy 设成 "marker", 插件才会向上找 projectRootMarkers(默认 .git)。 | 文件 | 职责 | 写入方式 | |-…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 1
- mo. downloads
- –
- Last push
- 2026-09-18
- Created
- 2026-09-12
Links
Basic safety check
- Findings
- curated 收录但无 npm 包/安装命令
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- agent-memory, cordis, cross-session-memory, deepseek-harness, dsh, dsh-plugin, markdown, memory
Related plugins
dsh-mnemon
omdsh-dev/dsh-mnemon
Cross-agent, local-first persistent memory plugin for DeepSeek Harness (DSH), powered by Mnemon. It shares long-term memory across Mnemon-enabled agents and adds runtime memory, searchable project documents, semantic recall, knowledge graph, and a Sidebar UI.
memsearch
zilliztech/memsearch
A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex, DSH), backed by Markdown and Milvus.
dsh-mnemon
dsh-external/dsh-mnemon
Composable, view-based memory for DeepSeek Harness. Pluggable sources and strategies, with three-tier memory out of the box.
markdownify-mcp
zcaceres/markdownify-mcp
A Model Context Protocol server for converting almost anything to Markdown