dsh-memory
Run the following command in DeepSeek Harness:
dsh plugin install haitang1/dsh-memory
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run the command: dsh plugin install haitang1/dsh-memory. Source: https://github.com/haitang1/dsh-memory
About this plugin
When working with DeepSeek Harness or Codex-like systems, a common pain point is the loss of context between sessions. The dsh-memory plugin addresses this by providing a durable, auto-injected memory layer for every session. Through its intelligent summarization mechanism, the plugin automatically distills conversation turns into summaries and periodically re-consolidates them into a global memory file, ensuring the agent never forgets previous interactions.
Its core capability lies in robust memory management tools. Users can interact with the memory using tools like memory_read, memory_add, and memory_search, while also benefiting from scoped management across workspaces or projects. Additionally, the plugin includes an auto-memory skill that guides agents to proactively identify key facts and write them to memory, enabling truly autonomous memory maintenance.
This is an ideal choice for developers building complex AI applications or agents that require long-term coherence and contextual memory. By injecting distilled summaries into every prompt, it ensures agents do not forget past interactions, significantly enhancing the quality and reliability of multi-turn dialogues.
Use Cases
- Build agent systems with long-term memory capabilities
- Ensure context is not lost in multi-turn dialogue
- Provide persistent knowledge storage for AI assistants
Best For
- DeepSeek Harness plugin developers
- AI engineers building complex RAG applications
- Agent project managers handling long-term interaction tasks
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.