Memory & Context
Memory, context storage, notes, and long-running state.
MindMemOS is a memory operating layer for AI agents. Its DeepSeek Harness plugin lets dsh agents automatically recall and write MindMemOS memories. It supports reusable memory across agents, self-evolving memory, and memory-to-skill workflows.
sivtr is a unified memory workspace for agents and humans, capturing terminal and AI session output into a shared local archive that can be searched and read via CLI, MCP, TUI, and Web API.
DSH插件:养一个会自己刷b站会成长的agent:不通过面板配置,通过学习诞生人格。
Cross-agent plugin for manually excluding, editing, and selectively compressing AI conversation context, with search, filtering, hide/restore, undo, and preservation of the original Session. Includes a DeepSeek Harness adapter.
Community memory plugin collection for DeepSeek Harness with 8 packages covering CJK full-text search, tool-result deduplication, hybrid recall, cross-session core memory, traceable compaction, and skill management. The README reports 130 automated tests and documents DSH 0.2.0-rc.2 Web-host, persistence, and cold-restart validation; user-pinned memories and skills use provenance records.
A third-party note-taking plugin for DeepSeek Harness (DSH) that provides a full Markdown notes manager and editor for capturing conversation content into notes and maintaining notes synced to a Git repository.
dsh-layered-memory is a DeepSeek Harness plugin for cross-session long-term memory. Its README describes background L0 capture, L1-L3 distillation, automatic recall injection, separate personal and work memories, SQLite storage, optional embeddings, and zero-config operation.
Theory repository for the Common Trust Protocol (Intelligentics v3.4). README states the repository holds the protocol's theoretical layer, while the DeepSeek Harness plugin implementation lives in the separate repository FuRongJun-1999/dsh-memory (npm: @furongjun1999/dsh-memory). It describes the Lingshu AEIS engineering layer: an MCP server for trustworthy agent memory and knowledge graphs, five-layer memory, condition-routed knowledge cards, an eight-step self-iteration loop, roleplay engine,
DeepSeek Harness plugin that connects over HTTP to a running PowerContext Server for recall, memory, handoff, experience, and skills. It does not embed storage, start the Server, or use MCP.
Co-Engram is a self-evolving team memory system for AI agents and teams. It runs as a native Cordis plugin for DeepSeek Harness, registers 40 tools, injects a memory prompt section, and shares a Markdown-in-Git data repo across Claude Code, OpenClaw, and DSH hosts via a process lock.
Session handoff and context management plugin for DeepSeek Harness, providing structured handoff export/resume, active context pruning via the compaction API, session management tools, and model route switching.
把一个目录里的每个 .md 变成能对话的 agent 人设
MemoryLake Harness provides client-side harnesses that connect coding agents to Memory Lake, a cross-device long-term memory service. Its DeepSeek Harness plugin is published as @memorylake/dsh-plugin and includes memory tools, prompt guidance, and session status; harnesses share one identity and data tree at ~/.memorylake/harness/.
住在屏幕彼端的ta(DSH 插件;纯聊天可脱离 DSH 独立运行)
An installable DeepSeek Harness (DSH) plugin package that imports conversation history from chat.deepseek.com into DSH sessions. Imported conversations appear in the left workspace list and can be opened, viewed, and continued with a selectable model. The package declares a dsh.bundle manifest (host half plus web client half) and is installed with `dsh plugin add`.
Global auto-memory plugin for DeepSeek Harness (DSH): it injects memory at session start, provides a memory_search retrieval tool, and includes compaction checkpoint archive reminders (described as a zero-write reminder approach).
dsh-hub is a DSH plugin hub that integrates global memory, graph-memory, plugin market, and self-update checks into one settings-page plugin.
User-authored tiered long-term memory for DeepSeek Harness: injected as a system-prompt section, with a Chinese panel to view and edit it. The AI never writes to memory on its own.
台账记忆(Ledger Memory)—— DSH 的工程化项目记忆:台账 + 活动日志(L0-L4) + 三层上下文压缩 + 逐字胶囊落盘 + 跨会话交接单。可回溯、可追查。
openclaw - skills : List · 智能表单(万能信息记录·智能记事本·智能账单)- Smart Form & Notes
dsh-headroom is a DeepSeek Harness plugin that compresses tool output before it enters model context. It routes JSON, search, log, tabular, and long-text content through dedicated compressors; lossy compressions store originals in a local CCR store and inject a marker for retrieval via headroom_retrieve(id=…).
DSH ExpMem is a community plugin for long-term personal memory in DeepSeek Harness. It stores habits, reusable task experience, and engineering insights in inspectable local JSON files, with ranked retrieval, reflection provenance, and a trust lifecycle.
HME (Harness-Memory-Evolution) is a cross-session long-term memory plugin for DeepSeek Harness. It provides USER, MEMORY, and archive layers, recall keyword retrieval, TTL and value tiers, plus a startup banner and hme-status tool/command. Current version v0.5.1; package @yinging/dsh-hme.
interest-memory is a lightweight, local-first long-term memory backend for AI agents. It runs as a single ~18MB binary with SQLite, extracts and verifies interest points at session end, and recalls relevant context at session start. Supports multiple agents, MCP, and graph-based memory traversal.