← 返回 Skills 市场
146
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install vc-shared-memory
功能描述
共享记忆池 - 跨团队、跨窗口、跨会话的统一记忆系统。支持公司公告板、项目协作空间、知识库、员工状态追踪。适用于虚拟公司、多人协作、AI 团队工作场景。
安全使用建议
This skill appears internally consistent for a local shared-memory/knowledge system: it stores JSON files under ~/.shared-memory and exposes a Node CLI and API to read/write/search that data. Before installing or running it: (1) review the full shared-memory.js (the provided excerpt is truncated) to confirm there are no network calls or hidden behavior; (2) run it in a sandbox or throwaway account if you want to inspect behavior first; (3) be aware it will create persistent files in your home directory (backups, audit logs) — consider file permissions, retention, and whether sensitive personal/company data will be stored unencrypted; (4) the SKILL.md mentions future notification integrations (Feishu / agent sessions_send) which, if enabled later, will require credentials — only provide such tokens when you trust the skill and have reviewed the notification code; (5) if you need stronger guarantees, request or add encryption-at-rest or access controls and audit the remaining (truncated) portions of shared-memory.js and test.js before use.
功能分析
Type: OpenClaw Skill
Name: vc-shared-memory
Version: 1.0.0
The shared-memory skill bundle implements a local, file-based persistence system for AI agents to share state across sessions and teams. It manages structured JSON data (announcements, projects, and knowledge bases) within a dedicated directory in the user's home folder (~/.shared-memory). The implementation in shared-memory.js is transparent, uses only standard Node.js modules (fs, path), and lacks any network connectivity, obfuscation, or instructions indicative of data exfiltration or prompt injection attacks.
能力评估
Purpose & Capability
Name/description (shared memory / collaboration) match the code and SKILL.md: the code implements read/write/search/update for announcements, projects, knowledge, employee status and related APIs and CLI commands. No unrelated credentials or binaries are requested.
Instruction Scope
The SKILL.md instructs the agent to read/write/manage records stored under ~/.shared-memory and to trigger notifications (future). The instructions are scoped to the stated purpose, but they require creating and modifying files in the user's home directory (privacy/persistence implication). SKILL.md also mentions future notification channels (Feishu, '子代理 sessions_send') which would expand scope if implemented.
Install Mechanism
No install spec — code is shipped in the package and is intended to run with node. No downloads or external installers are declared, which minimizes install-time risk. Running the provided Node CLI will create files under the home directory.
Credentials
The skill declares no required environment variables or credentials. The code uses HOME/USERPROFILE to determine ~/.shared-memory — this is proportional. Note: future notification integrations would likely require new credentials (not present now).
Persistence & Privilege
The skill persistently writes data to ~/.shared-memory (including audit logs and backups) and autoBackup is enabled by default. It does not set up as 'always' but has persistent on-disk presence and will modify the user's home filesystem; users should be aware of sensitive data retention and file permissions.
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 vc-shared-memory - 安装完成后，直接呼叫该 Skill 的名称或使用
/vc-shared-memory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
Shared Memory v1.0.0 introduces a unified memory system for cross-team collaboration.
- Provides a shared memory pool for company announcements, project collaboration, knowledge management, and employee status tracking.
- Supports scenarios such as virtual companies, multi-person teams, and AI-driven workflows.
- Includes modular storage structure for announcements, projects, knowledge base, employee status, and cross-team links.
- Offers core API and CLI tools for adding, updating, searching, and managing shared information.
- Enables automatic notification and indexing for efficient, real-time collaboration across teams and sessions.
元数据
常见问题
Shared Memory 是什么？
共享记忆池 - 跨团队、跨窗口、跨会话的统一记忆系统。支持公司公告板、项目协作空间、知识库、员工状态追踪。适用于虚拟公司、多人协作、AI 团队工作场景。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 146 次。
如何安装 Shared Memory？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vc-shared-memory」即可一键安装，无需额外配置。
Shared Memory 是免费的吗？
是的，Shared Memory 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Shared Memory 支持哪些平台？
Shared Memory 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 Shared Memory？
由 davidme6（@davidme6）开发并维护，当前版本 v1.0.0。
推荐 Skills