面向智慧體的蒸餾式長期記憶：依語意召回、依證據門控寫入，每種智慧體模式對應一個 kura。原生 DeepSeek Harness 外掛，無須向量資料庫、嵌入模型或框架。
DSH 整合
原生執行環境
作者聲明
安全稽核
未稽核
最後核實
2026-09-04
授權條款
MIT
01它能幫你完成什麼？
Give each DeepSeek Harness agent preset its own persistent memory store
A separate kura store per agent mode, recalled by meaning, so a preset change switches what the agent remembers
DSH users running multiple agent presets who want mode-scoped long-term memory
Write agent knowledge without poisoning the store
Memories distilled and poured only after an evidence gate (class, candidate, judge), not blind accumulation
Builders who want durable, auditable agent memory instead of unchecked assertions
02如何將外掛接入 DeepSeek Harness？
先決條件
- Python 3.11+ — no third-party dependencies (`[dev]` adds only pytest)
- Node.js 20+ — required only for the DSH plugin
- An OpenAI-compatible model endpoint answering `POST <url>/chat/completions`
安裝步驟
- 01
$ git clone https://github.com/lna-lab/distill-kura && cd distill-kura
- 02
pip install -e . # or just run: python3 -m distill_kura.cli
- 03
cp kura.example.toml kura.toml # edit: one model endpoint is enough to start
- 04
kura init main --path ~/kura/main # create an empty store
- 05
kura serve # http://127.0.0.1:8085
驗證整合成功
- curl -s -X POST localhost:8085/recall -H 'content-type: application/json' \
03DSH 整合程度與能力邊界
Loads as a native DeepSeek Harness plugin through `@deepseek-ai/dsh-tools` `defineTool`, exposing `kura_*` tools bound to a preset's memory store.
Semantic recall (recognition)
a question, or a named memory→picked memory slugs plus their [[link]] neighbourhood as context
Evidence-gated writing
a journal of agent events→new memories poured only after class → candidate → gate → judge
Per-mode memory stores
an agent preset / mode→a bound kura store; a preset change switches the active memory
Ships as DSH plugin, MCP server, HTTP service, Python library
any host→the same engine behind a DSH plugin, MCP, HTTP (:8085) or import
04適合誰？何時不該用？
適合
- DSH users running multiple agent presets who want mode-scoped long-term memory
- Builders who want durable, auditable agent memory instead of unchecked assertions
不適合
- Requires an OpenAI-compatible `POST <url>/chat/completions` model endpoint; the API key is read from an environment variable you name, never stored in config.
05相容性、維護與安全提醒
- The DSH plugin needs Node.js 20+; the core engine is Python 3.11+ with no third-party dependencies.
- Requires an OpenAI-compatible `POST <url>/chat/completions` model endpoint; the API key is read from an environment variable you name, never stored in config.
- Standard library only — no vector database, no embeddings, no framework; recognition uses a small local model over the whole index in one prompt.
Active repo (master), MIT licensed, no tagged release as of 2026-09-03.
06常見問題
如何把 distill-kura 安裝並接入 DeepSeek Harness？
它作為原生 DSH 外掛，透過 `@deepseek-ai/dsh-tools` 的 `defineTool` 載入，暴露綁定到預設倉庫的 `kura_*` 工具。在預設中指定倉庫，切換預設即切換整個記憶體。
沒有嵌入模型，召回是怎麼做到的？
整個索引（每條記憶一行）放進一次提示，由小模型辨識與問題相關的記憶；命中具名記憶則為確定性命中。README 稱約 500 條記憶的索引約 6k token。
記憶寫入安全嗎？
寫入受證據門控：日誌先分類、產生候選、通過 GATE，再經撰寫、草稿、評判，最後才倒入；不會把樸素斷言直接記成事實。
執行需要什麼環境？
Python 3.11+ 且無第三方依賴；僅 DSH 外掛需要 Node.js 20+；以及任意相容 OpenAI 的模型端點（`POST <url>/chat/completions`）。
07相關的 DSH 工作流程
weknora
作者 tencent
開源的 LLM 知識庫平臺：把原始文件轉化為可檢索的 RAG 知識庫、自主推理智慧體和自動維護的 Wiki 系統，支援多租戶部署。
honcho
作者 plastic-labs
用於構建有狀態 AI 智慧體的記憶庫，支援跨會話連續學習與上下文工程。
mirage
作者 strukto-ai
全球首個面向 AI 智慧體的統一虛擬檔案系統。
reme
作者 agentscope-ai
ReMe：面向 AI Agent 的記憶管理套件——幫 Agent「記住我、最佳化我」，支援 RAG 與長期記憶。
08資料與來源
Ships as a [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) plugin,
DSH switches **persona and tools** by agent preset. distill-kura switches **memory** by
此頁面根據專案公開文件、儲存庫中繼資料與 DSH Plugins 的結構化解析所產生；最後核實於 2026-09-04。發現錯誤？提交更正。
最佳 DeepSeek Harness 外掛
從全目錄挑出的 12 個值得優先安裝的外掛，涵蓋各個分類。