dsh-agentmemory
Run the following command in DeepSeek Harness:
dsh plugin install Yiipu/dsh-agentmemory
Paste the following prompt into your AI chat to install this plugin:
To install the dsh-agentmemory plugin in DeepSeek Harness, run: `dsh plugin install Yiipu/dsh-agentmemory`。. Source: https://github.com/Yiipu/dsh-agentmemory
About this plugin
DSH sessions are ephemeral, meaning once a session concludes, the agent's learned knowledge—including tool calls, recovered errors, and settled decisions—is permanently lost. This limitation hinders long-term improvement and context retention. dsh-agentmemory solves this by acting as a bridge between DSH and the local agentmemory daemon, persisting every session's activity into a searchable store. By mapping DSH events to the daemon's standard hooks, the plugin captures critical details like tool usage and errors, while its context injection capabilities allow the model to recall relevant memories during execution. Furthermore, it exposes explicit tools for curated memory management. This solution is perfect for developers building autonomous agents who need long-term context awareness, enabling their AI to utilize historical experience across sessions without being limited by single-turn context windows.
Use Cases
- Maintain agent knowledge accumulation across sessions
- Record tool call history and error logs
- Maintain long-term project-level context memory
Best For
- DSH developers needing long-term memory features
- AI engineers building complex agent applications
- Developers seeking to optimize model decision-making capabilities
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.