dsh-agent-memory: Long-Term Memory Plugin for DeepSeek Harness
The long-term memory plugin dsh-agent-memory for DeepSeek Harness (DSH) is designed to address the loss of conversational context and the inability to persist preferences. Featuring a zero-runtime-dependency architecture, it uses a hook mechanism to collect conversational evidence in real time and supports preference population through CLI. It provides a Web UI status panel, a data browser, and capability probing with circuit breaking. After installation, it should be used with the Python CLI to backfill data and record preferences, enabling dual-layer injection of global preferences and session memory.
Read Moredsh-agent-memory: Agent-driven long-term memory for DSH
dsh-agent-memory is an MIT-licensed long-term memory plugin (v0.3) built for DeepSeek Harness (DSH), aiming to provide cross-session structured memory capabilities. It adopts a hierarchical storage architecture, dividing memory into L1 facts (preferences and environment), L2 knowledge (reusable content), L3 contexts (timeline of experiences), and a summary layer, while supporting routing for both global and workspace scopes. The plugin exposes tools such as remember, recall, memory_browse, update/forget, and memory_stats to agents, supporting multi-dimensional retrieval and time pyramid browsing. Its key features include automatic time bucket compression and integration with dsh-agent-compact, automatically consolidating che...
Read More