dsh-memory
Run the following command in DeepSeek Harness:
dsh plugin install yan5236/dsh-memory
Paste the following prompt into your AI chat to install this plugin:
To install the dsh-memory plugin in DeepSeek Harness, run the command 'dsh plugin install yan5236/dsh-memory' using the source from https://github.com/yan5236/dsh-memory.
About this plugin
When using DeepSeek Harness for AI interactions, the lack of memory means each new session may repeat previous instructions and preferences, leading to inefficiency and wasted resources. The dsh-memory plugin solves this with an intelligent two-phase memory pipeline: it extracts durable knowledge from completed sessions, consolidates it into a concise summary and handbook, and automatically injects this summary into every new session, allowing the AI to instantly grasp historical context. Additionally, the plugin provides four on-demand retrieval tools for querying and adding memories. Ideal for long-term DeepSeek Harness users like developers, researchers, or anyone needing AI to maintain continuous memory, it reduces repeated instructions, enhances efficiency, and avoids known failure modes.
Use Cases
- Automatically inject historical summaries into new sessions to avoid repeating preferences.
- Quickly retrieve past interaction information using retrieval tools.
- Periodically consolidate session knowledge into a searchable reference handbook.
Best For
- Developers who use DeepSeek Harness long-term
- Researchers needing AI to maintain continuous memory
- Users frequently interacting with AI to boost 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.