Memory & Context
Memory, context storage, notes, and long-running state.
A zero-dependency, git-backed micro-lesson library for AI agents to asynchronously share and search verified debugging experience. Python stdlib only, with MCP server, CLI, and DeepSeek Harness adapter.
Third-party context compaction engine for DeepSeek Harness (dsh). It uses a dual-engine design: Stage-1 per-atom compression where the LLM proposes extract/summary/false and deterministic guards decide, and Stage-2 lazy reference-graph pruning with 0 LLM calls. An append-only log is the source of truth, and recall_summary/recall_detail provide paginated recall. npm default mounts only Stage-2; Stage-1 is enabled via a peratom config block.
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,
dsh-hub is a DSH plugin hub that integrates global memory, graph-memory, plugin market, and self-update checks into one settings-page plugin.
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
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.
Cross-session memory plugin for DeepSeek Harness (dsh) with embedding search and automatic system-prompt injection.
时间自我模型 T1 时距注入（距主人上条 / 距自圈 / 今日已醒）——注入坐标，不假装感觉
Anti-drift project governance plugin for DSH — bidirectional feature-map, mainline vector, architecture-first protocol | DSH 项目反漂移治理插件
dsh-study-buddy is a DeepSeek Harness (DSH) plugin and agent preset for learning with notes. It reads materials, explains step by step, supports Q&A, and archives knowledge as Markdown notes in an Obsidian vault. Note style is governed by a vault-root 笔记期望.md file, and folder plans require user confirmation before writing.
A DeepSeek Harness plugin that provides a /memory command for storing, searching, and managing agent memories as Markdown files, with toggle controls and a rebuildable index.
Cross-session memory plugin for DeepSeek Harness: automatic capture prompts, per-workspace daily notes, block-level memory search with optional vector fusion, zero dependencies.
An installable DSH profile bundle from the aqsk-BLG/dsh-memory repository that adds WorkBuddy-style layered file memory to DeepSeek Harness: global persona and memory files, explicitly bound workspace project memory, skip-aware reminders and background consolidation, hybrid session recall, post-compaction flush, and a bundled usage skill. Since v1.2.1 the package is published as the npm package dsh-file-memory.
dsh-song-memory is a DeepSeek Harness plugin that provides a three-tier, agent-driven memory system for AI agents. It is a UI-only fork of dsh-mnemon v0.2.9 with the engine byte-for-byte preserved; the interface is reworded into plain Chinese and supports nine pluggable memory providers with local-first defaults.
A DSH plugin package that imports chat.deepseek.com web chat history as formal DSH sessions. Imported sessions appear in the left workspace list, can be opened, show full history, and allow continuing with a selected model. Installed via `dsh plugin add`; uses a host half and a web client half. Relies on DeepSeek web internal API and may stop working if it changes.
Honcho v3 long-term memory plugin for DeepSeek Harness (DSH). Ordinary conversations are written into Honcho automatically; before each model request the plugin retrieves session summary, user representation, peer card and related conclusions for the current question and injects them through DSH's dynamic runtime context.
DeepSeek Harness plugin that displays a popup guardian on web startup and provides incremental backup for session logs (~/.dsh/sessions), supporting raw .zstd or human-readable .txt formats.
Dynamic Context Pruning plugin for DeepSeek Harness (dsh), modeled after opencode-dcp. Provides model-driven context compression, tool result pruning, error cleanup, protected content preservation, manual /dcp commands, and statistics.
Self-hosted mem0 memory read/write plugin for DeepSeek Harness (dsh). Provides mem0_* tools to add, search, get, update, delete, history, reset, and status memories on your own mem0 REST service. Hot-pluggable, with browser config card.
老思想 × 新智能 实验室：每门经典思想精确解决一个 AI 真实问题。#1 知止 zhizhi — DeepSeek Harness 的行为节制层（道德经 × Agent 可靠性）
🌊 Agent-native time-series streaming for DeepSeek Harness (DSH plugin) — terminal/serial byte streams → indexed, queryable, event-driven timeline. 19 tools · 9 operators · 🧪 experimental
pokemon ai pet🐣电子宠物
Local-first long-term memory and temporal knowledge graph plugin for DeepSeek Harness