omdsh-dev / dsh-mnemon

LLM-supervised persistent memory for AI agents — graph-based recall, cross-session knowledge, single binary. Works with DeepSeek Harness, Claude Code, OpenClaw, and any agent runtime.

Local-first AI workbench for DSH Plugins, combining Agent sessions, project files, data analysis, web research, MCP, and Office artifacts in an Electron desktop app.

语织：中文优先、用户可控的 AI 写作风格记忆层，让 AI 学会你的表达，需要时再调用。Chinese-first style memory for AI; DeepSeek Harness adapter included.

Open-source CMA-compatible agent runtime for any model, with MCP tools, sandboxed sessions, audit, replay, and a local console. Includes a native DeepSeek Harness bundle over stdio MCP.

Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness (capability seam: ctx.memory + SQLite provider + memory tool + frozen snapshot injection)

MemoryHub (mh) plugin for DeepSeek Harness (dsh): auto-loads checkpoint memory on session start, adds mh_* tools and the mh skill, and a Memory tab in the web UI