dsh-light-memory
Run the following command in DeepSeek Harness:
dsh plugin install chidaic/dsh-light-memory
Paste the following prompt into your AI chat to install this plugin:
Visit https://github.com/chidaic/dsh-light-memory to obtain the plugin, and install it using the command dsh plugin install chidaic/dsh-light-memory, then restart the DeepSeek Harness Web GUI to take effect.
About this plugin
As AI assistants become integral to productivity, memory management often gets bogged down by complex databases or vector stores. The dsh-light-memory plugin offers an ultra-lightweight solution for DeepSeek Harness users, leveraging four simple Markdown files—USER.md, PROJECT.md, WORKLOG.md, and CONVENTION.md—to build a complete persistent memory system. These plain text files are editable, version-controllable via git, and keep all data fully in the user's hands without any external dependencies.
At the heart of the plugin are two key actions: append automatically adds operation logs to WORKLOG.md, ensuring a clear and traceable workflow history, while distill intelligently analyzes session content at task boundaries, distilling important conclusions into user-level or project-level files. Users can customize memory conventions by editing CONVENTION.md, gaining flexible control over what and how to record, embodying a truly customizable and pluggable approach. Built-in security features like guard limits and secret scanning further protect memory data from misuse or leakage.
dsh-light-memory is ideal for all DeepSeek Harness users who value simplicity, transparency, and control over their memory management—whether individual developers or team collaborators. It makes memory visible and manageable, helping enhance AI assistant consistency without adding overhead. With its zero-database, zero-telemetry design, it stands out as a perfect choice for lightweight AI memory systems.
Screenshots
Use Cases
- Automatically log operation flow to a log file after AI sessions.
- Intelligently extract key conclusions at task boundaries and distill them into persistent files.
- Customize memory methods and content by editing convention files.
Best For
- Developers and users of DeepSeek Harness.
- Individuals or teams needing transparent, editable memory management.
- Collaborators seeking lightweight tools with zero external dependencies.
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.