dsh-memory-system
Local-first persistent memory for DeepSeek Harness: bounded pre-step hot-memory injection, exact and Chinese-bigram BM25 cold recall, six agent tools, approval-gated lease-lock writes with recovery receipts, and read-only governance and trajectory-review candidates.
面向 DeepSeek Harness 的本地优先持久记忆:在 pre-step 注入有预算上限的热记忆,使用精确匹配与中文二元 BM25 召回冷层,提供六个 Agent 工具,并以审批门禁、租约锁、恢复回执和只读治理与轨迹复盘候选保护写入。
How to install
dsh plugin add @zhujunpeng12/dsh-memory-system About
dsh-memory-system **让 DeepSeek Harness 跨会话记住项目、规则和纠正,同时把数据留在本机 Markdown。** English · 工作原理 · 安全边界 · 参与贡献 它不是另一个黑盒向量库:启动时只注入有预算的热记忆,需要历史细节时才做可解释的中文 BM25 召回;所有写入默认预览,并通过租约锁、before-image、SHA-256 前置条件和回执保护。默认使用 ~/.dsh-memory/,无需 Obsidian、数据库、向量服务或外部 API。 **隐私边界**:仓库只包含机制,不包含任何个人数据。画像、规则、事件和项目笔记始终留在使用者本机;也可用 MEMORY_VAULT 指向自己的 Obsidian Vault。 30 秒安装 前提:已安装 DeepSeek Harness 0.1.0-rc.7,Node.js 22/24 与 Python 3.10+ 可用。 bash npx @deepseek-ai/dsh plugin --profile web add github:zhujunpeng12/dsh-memory-system 重启 Harness 后,在新会话让 Agent 运行 memory_gate。成功时会看到门禁结果,并且首轮上下文包含 [vault-bootstrap];首次运行会自动创建 ~/.dsh-m…
Recommendation signals
Meta
- License
- MIT
- Language
- TypeScript
- GitHub stars
- 8
- mo. downloads
- –
- Last push
- 2026-08-21
- Created
- 2026-08-15
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- agent-memory, ai-agents, chinese-bm25, coding-agent, deepseek-harness, dsh, dsh-plugin, local-first, memory-system, persistent-memory, plugin-evolution
Related plugins
hindsight
vectorize-io/hindsight/tree/main/hindsight-integrations/coding-agents
Hindsight, agent memory that learns: long-term project memory with auto recall and retain, knowledge pages, deep reflection, and per-repo memory banks.
brightdata-mcp
luminati-io/brightdata-mcp
A powerful Model Context Protocol (MCP) server that provides an all-in-one solution for public web access.
memsearch
zilliztech/memsearch/tree/main/plugins/dsh
Shared Markdown memory for DSH and other coding agents, with automatic capture, pre-step context injection, searchable recall, and memory-to-skill self-evolution through a review panel.
hivemind
activeloopai/hivemind
Hivemind turns your traces into reusable skills across agents