dsh-cc-haha-memory
Run the following command in DeepSeek Harness:
dsh plugin install yihefeikong-rgb/dsh-cc-haha-memory
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install yihefeikong-rgb/dsh-cc-haha-memory in your terminal to install the plugin; the source is available at https://github.com/yihefeikong-rgb/dsh-cc-haha-memory .
About this plugin
Every DeepSeek Harness session starts with a blank context window. Preferences you stated last week, project conventions, debugging conclusions—all gone the moment the session ends. dsh-cc-haha-memory fills that gap with a persistent, file-backed memory layer: when a user says "remember this," the plugin triggers a visible tool call that writes a real Markdown file to disk, so memory becomes a searchable, editable asset instead of an ephemeral prompt.
The design rests on three pillars. First, isolated file storage: each memory is its own Markdown file with a short index in MEMORY.md; updates target explicit memory IDs and preserve history, eliminating silent overwrites from fuzzy title guesses. Second, scoped context injection: each turn loads only global plus current-project memories, recalling at most five relevant entries based on the question, rather than flooding the context with every project at once. Third, automation and a Web UI: a background reviewer evaluates each conversation turn at a configurable interval for information worth keeping long-term, and a "Memory" tab at the top of the session page provides search, create, edit, and delete, plus non-destructive import from existing Claude Code memories.
Ideal for developers running multiple projects in DSH who want context to carry across sessions without manually pasting background notes each time. The plugin runs entirely on native DSH events and tool interfaces—no extra services required—so it is ready to accumulate memories as soon as DSH is restarted after installation.
Use Cases
- Carry over project conventions and preferences across sessions without re-pasting context
- Isolate memories per project to prevent cross-project context pollution during parallel work
- Migrate existing Claude Code memories into the DSH workflow without data loss
Best For
- Developers maintaining multiple long-term projects in DSH
- Teams that rely on automatic context recall across sessions instead of manual prompt repetition
- Individuals or small teams migrating from or running alongside Claude Code
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.