Noelune

Unified Agent Memory — Development skill for Claude Code

Development community

Unified fleet-wide agent memory system for DeepSeek Harness — shared Obsidian Vault for dsh, Codex, Claude Code & Hermes with dependency-free Python FTS5 core.

How to install Unified Agent Memory

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Noelune/unified-agent-memory and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Unified Agent Memory does

Unified fleet-wide agent memory system for DeepSeek Harness — shared Obsidian Vault for dsh, Codex, Claude Code & Hermes with dependency-free Python FTS5 core.

Alternatives in Development

IPolloWork — Enterprise-grade, local-first Agent Workbench for people and agent teams 4.9k ★
Petdex — A public gallery of animated pets for Codex, Claude Code, DeepSeek Harness, Hermes, OpenCode, Gemini CLI, and 4k ★
Claudes C Compiler — Claude Opus 4.6 wrote a dependency-free C compiler in Rust, with backends targeting x86 (64- and 32-bit), ARM 2.6k ★

README

🧠 unified-agent-memory

Unified Fleet-Wide Agent Memory System for DeepSeek Harness & Multi-Agent Runtimes

*跨 Agent 统一持久化记忆系统 — 基于 Obsidian Vault 存储规范与 SQLite FTS5 本地索引的知识生命周期管理基座*

[![npm version](https://img.shields.io/npm/v/dsh-unified-agent-memory?style=flat-square&color=c084fc)](https://www.npmjs.com/package/dsh-unified-agent-memory) [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg?style=flat-square)](LICENSE) [![Python Stdlib](https://img.shields.io/badge/python-3.9+%20stdlib--only-green.svg?style=flat-square)](#) [![Obsidian Native](https://img.shields.io/badge/storage-Obsidian%20Vault-purple.svg?style=flat-square)](#) [![Search Engine](https://img.shields.io/badge/index-SQLite%20FTS5-blue.svg?style=flat-square)](#)

[产品定位与问题定义](#-产品定位与问题定义) • [核心技术优势](#-核心技术优势) • [Obsidian 存储规范](#-obsidian-vault-存储结构规范) • [Agent 全流程自动部署](#-agent-全流程自动部署流程) • [生命周期控制流](#-知识生命周期控制流) • [安全隔离与防护机制](#-安全隔离与凭据脱敏机制)

📌 产品定位与问题定义

传统 Agent 记忆插件多数绑定于特定的 Agent 运行时（Single-Agent Scope），其存储空间与会话相互隔离，缺乏统一的状态持久化与跨代理知识共享能力。

**`unified-agent-memory` 架构旨在为多 Agent 舰队提供统一的知识管理基座**：它允许 `dsh`、`Codex`、`Claude Code` 与 `Hermes` 等多个独立代理共享基于 Markdown 规范的 Obsidian Vault 知识仓库，配合纯 Python 标准库核心与 SQLite FTS5 本地索引，构建涵盖**知识摄取 (Ingestion)、知识晋升 (Promotion)、冲突裁决 (Adjudication) 与衰减遗忘 (Decay/Retention)** 的全闭环生命周期管理系统。

核心与 DSH 适配器分层维护，具体职责和写入边界见 [docs/DSH-MEMORY-ADAPTERS.md](docs/DSH-MEMORY-ADAPTERS.md)。该文档明确区分 `unified-agent-memory`、`dsh-hermes-memory` 与 `dsh-memory-discipline`。

🚀 核心技术优势

统一单一真理来源 (Shared Source of Truth)
全 Fleet 代理共享一致的 Markdown 知识库规范。开发者与用户可直接借助 Obsidian 等编辑器可视化审计、人工校验或修改代理积累的知识节点。
纯 Python 标准库核心 (Zero-Dependency Engine)
core/ 模块完全基于 Python 标准库（sqlite3, json, hashlib, argparse, os）实现，无外部依赖开销，具备毫秒级冷启动性能。
SQLite FTS5 本地全文索引 (Local-First High-Performance Search)
检索索引持久化存储于本地 ~/.unified-memory/index-.db，按 Vault 隔离；基于 SQLite FTS5 引擎实现毫秒级全文匹配与相关度计算，数据隐私完全归属于本地宿主机。