A distilled long-term memory for agents: recall by meaning, writing gated by evidence, one kura per agent mode. Native DeepSeek Harness plugin with no vector database, embeddings, or framework.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-09-04
License
MIT
01What can it help you accomplish?
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
02How to install into DeepSeek Harness
Prerequisites
- 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`
Installation steps
- 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
Verify the integration
- curl -s -X POST localhost:8085/recall -H 'content-type: application/json' \
03DSH integration and capability boundaries
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
04Who is it for? When not to use it?
Good for
- DSH users running multiple agent presets who want mode-scoped long-term memory
- Builders who want durable, auditable agent memory instead of unchecked assertions
Not for
- 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.
05Compatibility, maintenance and safety notes
- 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.
06Frequently asked questions
How do I connect distill-kura to DeepSeek Harness?
It loads as a native DSH plugin through `@deepseek-ai/dsh-tools` `defineTool`, exposing `kura_*` tools bound to a preset's store. Name a store in the preset and one preset change moves the whole memory self.
How does recall work without embeddings?
The whole index (one line per memory) goes into one prompt and a small model recognises what bears on the question; a named memory is a deterministic hit. The README reports an ~500-memory index is about 6k tokens.
Are memory writes safe?
Writing is gated by evidence: a journal is classed, turned into candidates, passed through a GATE, then composed, drafted, judged, and only then poured. A naive assertion is not recorded as a fact.
What are the runtime requirements?
Python 3.11+ with no third-party dependencies, Node.js 20+ for the DSH plugin only, and any OpenAI-compatible model endpoint answering `POST <url>/chat/completions`.
07Related DSH workflows
weknora
by tencent
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
honcho
by plastic-labs
Memory library for building stateful agents
mirage
by strukto-ai
The World's First Virtual Terminal for AI Agents
reme
by agentscope-ai
ReMe: Memory Management Kit for Agents - Remember Me, Refine Me.
08Data and sources
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
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-09-04. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.