← 返回 Skills 市场
105
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install personal-memory-system
功能描述
个人AI记忆系统（Personal AI Memory System）。当用户说以下任何内容时立即触发： 记忆系统、记录今天、写日记、今天发生了、随手记、快速记录、 回顾过去、整理我的经历、帮我做决策、沙盘推演、这件事该不该做、 分析我最近的状态、预测一下、我的目标、追踪进度、生成周报、 生成月报、生成年报、这周...
安全使用建议
This skill appears internally consistent and is instruction-only (no code or network calls embedded). Before installing: 1) Verify where the memory directory will live — prefer a user-owned path (e.g., ~/memory or ~/.memory) rather than a root-level /memory to avoid permission issues. 2) If you care about privacy, run the skill with a local model or a platform that offers strict data-retention/no-retention guarantees; otherwise journal text you send to the model will be transmitted as conversation context to that provider. 3) Follow the skill's advice: set directory permissions (chmod 700), add the folder to .gitignore, and consider encrypting backups. 4) Be mindful when importing old content or writing about third parties — the system records what you give it. 5) The only minor red flag is inconsistent path notation in the docs; double-check install/initialization steps so files are created where you expect. If the package later includes code, network calls, or requests credentials, re-evaluate — that would change this assessment.
能力标签
能力评估
Purpose & Capability
The skill's name and description (personal AI memory / journaling / decision advisor) match the instructions: read/write local memory files, generate responses, run analyses, and produce reports. It does not ask for unrelated credentials, binaries, or external services, so requested capabilities are proportionate to the stated purpose.
Instruction Scope
The runtime instructions direct the agent to read and write many files inside a local memory directory and to include user content as model context when producing responses — that is expected for a memory/journal system. Two notes: (1) the SKILL uses inconsistent path forms ("/memory/", "memory/", and references to "~/.memory/"), which could cause accidental placement of data in an unintended location (e.g., a root-level /memory requiring elevated permissions); (2) the skill repeatedly instructs loading whole logs (e.g., 'signals-log 全部' for some tasks), which is coherent with analysis use but increases the volume of personal data the model will see — users should be aware that large historical context will be sent to whatever model service they use.
Install Mechanism
This is instruction-only with no install spec and no code files executed by the platform. That minimizes install-time risk; nothing is downloaded or extracted by the skill itself.
Credentials
The skill declares no required environment variables, no credentials, and no config paths outside its own memory directory. That is proportionate for a local journaling/analysis skill.
Persistence & Privilege
The skill does not request 'always: true' and sets autonomous_invocation: false; it asks the agent to store data in a local memory directory it manages itself. That level of persistence and privilege is appropriate for a personal-memory skill.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install personal-memory-system - 安装完成后，直接呼叫该 Skill 的名称或使用
/personal-memory-system触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v3.1.0
Version 3.1.0
- Added AGENTS.md, LICENSE.md, and README_EN.md for improved documentation and user guidance.
- Updated SKILL.md with a comprehensive safety & privacy statement, detailing local data storage, platform telemetry, and content handling principles.
- Enhanced and clarified file structure, user intent mapping, and context loading rules.
- Refined core rules (“铁律”) and AI character standards for transparency and user trust.
- Improved instructions on file permission and encryption for local privacy protection.
v1.2.0
**Version 1.2.0 – World-class upgrade for personal-memory-system**
- Introduced a comprehensive new system structure: reorganized into present, past, and future memory dimensions, each with dedicated files and folders for diaries, signals, goals, decisions, and more.
- Implemented precise conversation type recognition with tailored processing and file access rules for all supported memory, record, and decision scenarios.
- Enforced strict separation of user diary entries and AI responses for data integrity and privacy.
- Added context-sensitive data loading rules for each interaction to optimize relevance and performance.
- Defined “iron rules” for diary sanctity, decision storage, and empathetic AI response to user input.
- Refined the AI character role to act as a warm, direct, and honest long-term companion, not merely an assistant.
元数据
常见问题
personal-memory-system 是什么？
个人AI记忆系统（Personal AI Memory System）。当用户说以下任何内容时立即触发： 记忆系统、记录今天、写日记、今天发生了、随手记、快速记录、 回顾过去、整理我的经历、帮我做决策、沙盘推演、这件事该不该做、 分析我最近的状态、预测一下、我的目标、追踪进度、生成周报、 生成月报、生成年报、这周... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 105 次。
如何安装 personal-memory-system？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install personal-memory-system」即可一键安装，无需额外配置。
personal-memory-system 是免费的吗？
是的，personal-memory-system 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
personal-memory-system 支持哪些平台？
personal-memory-system 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 personal-memory-system？
由 ChenChen（@chenchen913）开发并维护，当前版本 v3.1.0。
推荐 Skills