A file-only persistent-memory layer for DSH that carries typed context between sessions.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-09-29
License
MIT
01What can it help you accomplish?
Retain durable user, project, feedback, and reference context across DSH sessions
Typed Markdown memory files and an automatically injected persistent-memory index
DSH users who want file-based recall without external services
02How to install into DeepSeek Harness
Prerequisites
- `@deepseek-ai/dsh >= 0.1.5-rc.2` and Node `^22.19 || >=24`
Installation steps
- 01
Run `dsh plugin --profile demo add dsh-auto-memory`.
$ dsh plugin --profile demo add dsh-auto-memory
- 02
Restart the profile with `dsh --profile demo`.
$ dsh --profile demo
Verify the integration
Not specified by the author
03DSH integration and capability boundaries
Native persistent-memory plugin that injects a typed Markdown memory index into DSH system prompts.
Typed memory tools
Model memory requests→`memory_write`, `memory_read`, `memory_list`, `memory_delete`, `memory_prune`, and `memory_delete_all`
Memory files are stored under `$DSH_HOME/memory/`.Automatic consolidation
A completed root session when `autoSummarize: true` is enabled→A background pass that extracts durable facts and files them as memories
04Who is it for? When not to use it?
Good for
- DSH users who want file-based recall without external services
Not for
- The README requires DSH `>= 0.1.5-rc.2` and Node `^22.19 || >=24`.
05Compatibility, maintenance and safety notes
- The README requires DSH `>= 0.1.5-rc.2` and Node `^22.19 || >=24`.
- Budget truncation is positional rather than relevance-ranked; the README presents full relevance ranking as future work.
MIT repository; facts captured from the README.
06Frequently asked questions
What does dsh-auto-memory persist?
It stores typed memories as ordinary Markdown files under `$DSH_HOME/memory/` and injects an index into DSH system prompts.
How do I install it?
Run `dsh plugin --profile demo add dsh-auto-memory`, then restart the profile with `dsh --profile demo`.
Does it need an external database or embedding service?
No. The README describes it as file-only, with no external services, databases, or embeddings.
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
webnovel-writer
by lingfengqaq
A long-form Chinese web-novel writing assistant for Claude Code and DeepSeek Harness, designed to reduce memory loss and hallucinations in serialized projects of up to 2 million characters.
mirage
by strukto-ai
The World's First Virtual Terminal for AI Agents
08Data and sources
`dsh-auto-memory` closes that gap natively:
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-09-29. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.