dsh-memory
Run the following command in DeepSeek Harness:
dsh plugin install Towzai/dsh-memory
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, install the dsh-memory plugin using the command `dsh plugin install Towzai/dsh-memory` from the source at https://github.com/Towzai/dsh-memory.
About this plugin
Managing memory in intelligent dialogue systems is a critical challenge. Traditional approaches often inject full memory history in every conversation turn, leading to token waste and LLM cache invalidation, which degrades response speed and cost efficiency. The dsh-memory plugin is a cross-session memory solution designed to address these issues.
Built on YAML storage, it offers six core tools: remember, recall, view, list, forget, and pin. It supports embedding semantic search combined with keyword matching for precise retrieval. The innovative 'title-first injection' strategy injects only concise titles in the first turn and expands details on demand, keeping prompts compact without breaking prefix caching. Two-level automatic injection (session-persistent and dynamic retrieval) allows the agent to inherently carry memory without explicit tool calls.
dsh-memory is ideal for developers using the DeepSeek Harness framework, especially those seeking to equip agents with long-term memory, optimize dialogue efficiency, and reduce token costs. Whether building personal assistants or complex interactive systems, it significantly enhances intelligent memory management and resource utilization.
Use Cases
- Building intelligent assistants with long-term memory
- Optimizing token usage in multi-turn dialogues
- Persisting information across sessions
Best For
- Developers using the DeepSeek Harness framework
- Technologists building intelligent dialogue systems
- Engineers focusing on AI application costs and efficiency
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.