DSH Plugin HubDSH 插件市场
AskTheWay/dsh-auto-memory 预览 preview

AskTheWay/dsh-auto-memory

Plugin插件 Native原生 ⭐ 75 MIT Memory & Knowledge记忆与知识库Prompts & Skills提示词与技能

Claude Code-style auto-memory plugin for DeepSeek Harness (dsh): typed memory files + MEMORY.md index auto-injected into the system prompt. File-only, no external services.

Project Overview项目介绍

Edited by Zhiwei Agent (zhiwei)由知微Agent(zhiwei)编辑

dsh-auto-memory is a native persistent-memory plugin built exclusively for DeepSeek Harness (DSH), installed into a chosen profile with a single command such as dsh plugin --profile demo add dsh-auto-memory and activated on the next restart of that profile. It exposes six typed tools — memory_write, memory_read, memory_list, memory_delete, memory_prune, and memory_delete_all — that store memories as ordinary Markdown files under $DSH_HOME/memory/, classified as user, feedback, project, or reference, and scoped across workspace and user layers to prevent cross-project leakage. A dynamic system-prompt section (order 4000) is re-evaluated on every step, byte-budgeted, and injects zero bytes when the store is empty, so Claude Code-style recall works without external services, embeddings, databases, or servers.

In a typical workflow the model is told "Remember: I'm a Python backend engineer preparing for interviews"; tomorrow, in a fresh session, asking "what do you know about me?" causes memory_read to resolve [[name]] cross-links one level deep and surface the stored summary. When autoSummarize: true is set, a silent background LLM pass at root-session end extracts durable new facts, deduplicates them, and files them as memories. The target audience is any individual or team that relies on DSH for long-running coding sessions and wants automatic, hands-off context recovery comparable to Claude Code without configuring third-party MCP servers.

Dependencies are minimal: @deepseek-ai/dsh >= 0.1.5-rc.2, Node ^22.19 || >=24, and exactly one runtime package, yaml, with an installed footprint of roughly 15 kB; the license is MIT. Configuration is wholesale-replaced under the profile's cordis.patch.yml, exposing maxBytes (default 4096), memoryDir (default $DSH_HOME/memory), and enableUserScope. First-run caveats: source installs need npm install && npm run build before dsh plugin add /abs/path; bulk memory_delete_all requires explicit human approval via tools/pre-execute; budget truncation is currently positional rather than relevance-ranked, so pinned probes retain at least 80% at half budget while ordinary probes degrade as N grows — full relevance ranking remains on the roadmap.

dsh-auto-memory 是一款面向 DeepSeek Harness（DSH）的原生持久记忆插件，以 dsh plugin add dsh-auto-memory 命令安装到指定 profile（例如 demo）后即可生效。它提供 memory_write、memory_read、memory_list、memory_delete、memory_prune 与 memory_delete_all 六个工具，将记忆以普通 Markdown 文件的形式落地于 $DSH_HOME/memory/ 目录，并按 user / feedback / project / reference 四类组织，支持 workspace 与 user 双层作用域以避免跨项目泄漏。每次模型请求都会重新评估系统提示中的索引区，并在内容为空时注入 0 字节。

典型用法是让代理在日常对话中调用 memory_write 记录偏好与项目背景，并在新会话中通过 memory_read 触发 [[name]] 交叉链接展开，使上下文自动恢复；启用 autoSummarize 后，根会话结束时会异步抽取持久事实做去重归档。插件还内置了过零读取的软隐藏淘汰、由 tools/pre-execute 人工确认的批量删除防护，以及 deterministic eval 层保障的字节预算、锁定自愈与 {{ 注入清洗。它适合任何长期使用 DSH、又希望拥有类 Claude Code 自动记忆体验的个人或团队。

依赖方面，仅要求 @deepseek-ai/dsh >= 0.1.5-rc.2 与 Node ^22.19 || >=24，运行时仅额外引入 yaml 一个依赖，安装体积约 15 kB，许可证为 MIT。安装时若通过源码需执行 npm install && npm run build && dsh plugin --profile demo add /abs/path，并可通过 cordis.patch.yml 覆盖 maxBytes（默认 4096）、memoryDir、enableUserScope 等配置。已知限制：当前按索引顺序截断，相关性排序尚未实现，pinned 探针在半数预算下保留率可达 80% 以上，普通探针仍会随记忆数下降。

View on GitHub在 GitHub 查看

Recommended强烈推荐 Send to your DSH发给你的 DSH 安装 
请帮我安装这个 DSH 插件。安装前先完成【兼容性检查 + 安全性检查】，检查通过再动手。

插件：dsh-auto-memory（AskTheWay/dsh-auto-memory）
仓库：https://github.com/AskTheWay/dsh-auto-memory
本站详情页：https://www.yhbd.top/plugins/asktheway-dsh-auto-memory/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 75 · 最近提交 2026-09-28 · 主语言 TypeScript

按下面顺序执行，每步先把结论告诉我，再进入下一步：

【1 兼容性检查】
① 我这边：DSH 版本、Node 版本、操作系统、当前 profile（web / desktop）。
② 读它的 README、package.json、插件 manifest，列出它要求的 DSH 版本 / Node 版本 / 操作系统 / 外部依赖 / 需要另外先装的运行时。
③ 逐条比对，结论只写「满足 / 不满足 / 未知」三种；不满足的给出可行替代方案。
④ 检查是否和我已装的插件冲突：命令名重复、skill / tool 重名、端口占用、重复注册的 MCP server。

【2 安全性检查】
① 仓库可信度：和上面「本站登记」是否一致；star / fork 数、创建时间、最近提交，是否归档或长期停更。
② 安装脚本：逐行看 package.json 的 preinstall / install / postinstall，以及 install.sh、setup.ps1 之类脚本。出现 curl|bash、下载后直接执行、混淆代码、访问与插件功能无关的域名，立刻停下来告诉我，不要继续装。
③ 依赖：列出新增依赖，标出无人维护、或与知名包拼写近似的可疑包（typosquatting）。
④ 权限与副作用：它会读写哪些目录、访问哪些域名、需要哪些 DSH 权限（filesystem / network / shell / clipboard 等），以及怎么卸载和回滚。
⑤ 如果它要求 sudo / 管理员权限，或权限明显超出功能所需，先停下来问我。

【3 安装】
上面两步没有「不满足」和「高危项」时才执行；用官方推荐方式安装，不要自行提权。

【4 汇报】
用表格输出：检查项 / 结论 / 依据 / 是否需要我决策。拿不准的一律写「未知」并说明要我怎么确认——不要猜，也不要替我决定。

Send this message to DSH in your current session: it verifies compatibility and security first (answering met / not met / unknown item by item) and only installs once everything checks out — it will stop and ask you if it finds a high-risk item. The box scrolls; the copy is the full prompt. CLI install commands may not be accurate across systems, so DSH is the safer route.把上面这条消息直接发给当前会话里的 DSH：它会先核对兼容性与安全性（逐条给「满足 / 不满足 / 未知」），确认没问题再安装，有高危项会停下来问你。框内可滚动，复制到的是完整提示词；安装命令不一定准确，发给 DSH 更稳。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 no risk signal found未发现风险信号
  • This site's static screen found no obvious risk signal (stars, license, activity, manifest)本站静态筛查没发现明显风险信号（星标、许可证、更新活跃度、清单完整度）
DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

  • DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
  • External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时（Electron / Python / Docker 等）是否齐备
  • Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突：命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

  • Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致，是否归档或长期停更
  • Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
  • curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
  • Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
  • Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需（filesystem / network / shell / clipboard）
  • Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限，以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照，不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装（适合开发者）

CLI Install命令行安装 

dsh plugin --profile demo add dsh-auto-memory

把 AskTheWay/dsh-auto-memory 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

8.9 KB · 缓存于 2026-09-29 05:50:01 · 8.9 KB

dsh-auto-memory

CI npm version npm downloads License: MIT Node

English | 中文

Your dsh agent forgets everything you tell it. Every. Single. Session.

Fix it with one command. Claude Code-style persistent memory for DeepSeek Harness — native, zero servers, zero embeddings, zero setup.

dsh plugin --profile demo add dsh-auto-memory

Say "Remember: I'm a Python backend engineer preparing for interviews" today — open a brand-new session tomorrow, ask "what do you know about me?", and it remembers.

Memory management panel in the dsh Web UI

What's new in 0.3.0 (P2)

  • Pinned memories (pinned: true on memory_write): pinned entries lead the index, survive budget truncation, and are exempt from staleness eviction — a trust anchor the user controls.
  • Eval-driven fix: the injection budget now covers the whole section (index + guidance); it used to overshoot by ~800 bytes. Caught by the new deterministic evaluation layer on its first run.
  • Deterministic eval layer (evals/) in CI: injection budget curves, eviction zero-misfire, link-expansion bounds, and a signal-to-noise characterization — which pinned-priority truncation then improved from 38% → ≥80% probe retention under half-budget pressure. Same budget, better memories.

What's new in 0.2.0 (P1)

  • Auto-consolidation (autoSummarize: true): when a root session ends, a background LLM pass extracts durable new facts from the session and files them as memories — deduplicated, capped, fully silent on failure. Claude Code doesn't do this automatically.
  • Forgetting & eviction: every memory carries lifecycle metadata (created/updated/reads); memory_read counts references; staleAfterDays soft-hides zero-reference stale memories from the injected index (files kept); memory_prune lists (dry-run) or deletes aged memories.
  • Recall expansion: memory_read resolves [[name]] cross-links one level and attaches linked summaries.
  • memory_delete_all — guarded by tools/pre-execute human approval: the model cannot self-confirm irreversible bulk deletes.
  • Hardened by a second adversarial review (11 agents): single-lock clear (no concurrent-write escape), conditional index rebuild on touch (no O(N) amplification), session-start stale refresh, subagent capture cleanup, abortable consolidation.

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要，完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

渲染基于仓库 AskTheWay/dsh-auto-memory 的 README.md。 在 GitHub 查看原文 →

Related相关推荐

← 上一个 Prev dsh-novel-writer 下一个 Next custom-first-control-prompt →