dsh-memory
Run the following command in DeepSeek Harness:
dsh plugin install biubiu23333333/dsh-memory
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install biubiu23333333/dsh-memory in DeepSeek Harness to install this plugin; full source is available at https://github.com/biubiu23333333/dsh-memory .
About this plugin
DeepSeek Harness agent sessions are stateless by nature—every new conversation forgets all previously learned preferences, decisions, and lessons. dsh-memory injects a persistent three-layer memory system at the host level, giving agents truly continuous memory across sessions and workspaces.
The three layers are long-term memory (profiles, facts, decisions, projects) maintained by the agent, short-term memory where the host automatically appends raw transcripts at each turn end, and session episodes where an LLM condenses each turn into a one-line summary. Consolidation is fully host-driven, independent of agent awareness, with a deterministic fallback when no LLM is available. Built-in memory_read, memory_write, and memory_search tools are provided, and system-prompt guidance is injected so that every agent preset benefits without any extra configuration.
Ideal for developers who frequently switch workspaces and want agents to retain project context and personal preferences. Pure ESM, zero build, with dependencies resolved automatically at runtime by DSH—install once and it just works.
Use Cases
- Retain project context and key decisions across sessions
- Share a single memory store across multiple workspaces
- Let agents proactively search and write long-term memory
Best For
- DeepSeek Harness users who switch between projects frequently
- Developers who want agents to remember personal preferences
- Builders running multiple agent presets who need a unified memory layer
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
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.