← 返回 Skills 市场
450
总下载
0
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install mem0-1-0-0
功能描述
Intelligent memory layer for Clawdbot using Mem0. Provides semantic search and automatic storage of user preferences, patterns, and context across conversati...
安全使用建议
This skill appears to implement the memory functionality it claims, but there are important inconsistencies to resolve before installing: (1) It requires an OpenAI API key (OPENAI_API_KEY) even though the registry metadata declares no required env vars — supply and protect that key if you proceed. (2) The package includes Node scripts and package.json but provides no install instructions; you should run npm install (or otherwise ensure mem0ai and transitive deps are available) in a controlled environment or container. (3) The default USER_ID is hardcoded to 'abhay' and the skill writes data to ~/.mem0 — confirm you are comfortable with local persistent storage and change the default userId before use to avoid cross-user mixing. If you need higher assurance, ask the publisher for an explicit install spec and a corrected requirements list, review mem0ai's privacy/telemetry docs, and test the scripts in an isolated environment first.
功能分析
Type: OpenClaw Skill
Name: mem0-1-0-0
Version: 1.0.0
The skill is designed for a legitimate purpose (intelligent memory for a chatbot) and includes good security practices in its documentation (e.g., explicitly warning against storing secrets). However, it is classified as 'suspicious' due to potential vulnerabilities. Specifically, the `scripts/mem0-add.js` file directly uses `JSON.parse` on user-controlled input (`--messages` argument) without explicit sanitization, which could lead to unexpected behavior or denial of service if crafted maliciously. More critically, all memory operations (`mem0-add.js`, `mem0-search.js`, `mem0-list.js`, `mem0-delete.js`) accept a `--user` argument to specify the `userId`. If the OpenClaw agent environment does not strictly enforce the `userId` based on the current user's context, a malicious actor could potentially exploit this to access, modify, or delete memories belonging to other users, representing a logical access control vulnerability.
能力评估
Purpose & Capability
The skill's name, description, and code consistently implement a local memory layer (uses mem0ai, local vector store, OpenAI embedder/LLM). However registry metadata claims no required environment variables while SKILL.md and scripts expect OPENAI_API_KEY. That metadata omission is an incoherence that affects whether the skill can run and what secrets it needs.
Instruction Scope
SKILL.md and the scripts limit activity to semantic search, add/list/delete operations, and local storage under ~/.mem0. The code does not attempt to read arbitrary system files or contact unexpected endpoints (it uses the mem0ai client which, per config, will call OpenAI). It documents what to store and not store. This is within the stated memory-purpose scope.
Install Mechanism
There is no install spec even though package.json, package-lock.json, and Node scripts are included. That means the skill either expects the environment to already have Node deps (mem0ai) installed or will fail. The package-lock references many transitive packages (normal for npm), but absence of an install instruction is a usability/security gap that could cause unexpected behavior.
Credentials
The skill actually reads process.env.OPENAI_API_KEY (and optionally JSON_OUTPUT) but the registry lists no required env vars. Requesting an OpenAI API key is proportionate to the stated purpose, but the omission from declared requirements is misleading. Also USER_ID is hardcoded to 'abhay' in config, which could cause cross-user data mixing unless overridden.
Persistence & Privilege
The skill persists data locally under ~/.mem0 and creates an SQLite history.db — that's expected for a memory store and does not modify other skills or system-wide settings. always is false. Note the fixed history path and default USER_ID which may lead to persistent data on disk and potential accidental sharing between contexts.
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 mem0-1-0-0 - 安装完成后，直接呼叫该 Skill 的名称或使用
/mem0-1-0-0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
v1.0.0
Initial release of mem0: adaptive memory for Clawdbot
- Adds Mem0, an intelligent memory layer that learns and recalls user preferences, patterns, and conversational context.
- Supports semantic search and automatic memory extraction for dynamic, adaptive responses.
- Provides commands for searching, adding, listing, and deleting memories via scripts.
- Complements structured MEMORY.md with learned, behavioral context.
- Delivers improved performance: higher accuracy, faster retrieval, and reduced token usage.
- Includes detailed storage guidelines and programmatic access options.
元数据
常见问题
Mem0 1.0.0 是什么？
Intelligent memory layer for Clawdbot using Mem0. Provides semantic search and automatic storage of user preferences, patterns, and context across conversati... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 450 次。
如何安装 Mem0 1.0.0？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install mem0-1-0-0」即可一键安装，无需额外配置。
Mem0 1.0.0 是免费的吗？
是的，Mem0 1.0.0 完全免费（开源免费），可自由下载、安装和使用。
Mem0 1.0.0 支持哪些平台？
Mem0 1.0.0 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 Mem0 1.0.0？
由 Sieyer（@sieyer）开发并维护，当前版本 v1.0.0。
推荐 Skills