dsh-self-improved
Run the following command in DeepSeek Harness:
dsh plugin install madage/dsh-self-improved
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, please visit https://github.com/madage/dsh-self-improved to obtain the official source code and follow the setup guide for local deployment.
About this plugin
The most common frustration when using AI assistants is the “start-from-scratch” syndrome, where traditional session isolation prevents models from building a lasting understanding of their users. dsh-selfimproved addresses this exact gap by enabling cross-session memory injection and continuous self-evolution directly within the DeepSeek Harness environment, ensuring your AI truly remembers you and adapts intelligently over time.
Built on a fully local four-tier memory pipeline, the plugin automatically distills key facts, preferences, and behavioral instructions from your dialogues into a private SQLite store. Before each new turn, it leverages hybrid vector-and-keyword retrieval to inject the most relevant context back into the conversation. Beyond basic recall, it features active self-evolution: memories are consolidated and aged-out information is pruned, successful interaction patterns are compiled into reusable skills, and your assistant’s persona continuously refines itself through real-world usage. Everything runs 100% offline with zero external data transmission.
Designed for creators seeking long-term digital companions, professionals aiming to eliminate repetitive briefings, and tinkerers building highly personalized AI ecosystems, this plugin integrates effortlessly into existing setups. With a built-in memory browser, granular control panels, and an efficient background maintenance schedule that keeps operational costs minimal, it empowers you to cultivate a deeply customized AI companion that learns, evolves, and grows alongside you.
Use Cases
- Track personal preferences and history across sessions for seamless conversation continuity
- Automatically distill daily work successes into standardized processes and reusable skills
- Run securely in strictly isolated network environments to completely prevent sensitive data leakage
Best For
- Users dedicated to building AI assistants with long-term memory and dynamic personalities
- Tech enthusiasts prioritizing data sovereignty and requiring fully local model/plugin deployments
- Knowledge workers relying on AI for complex projects who need knowledge retention and experience reuse
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.