← 返回 Skills 市场
Elite Longterm Memory 1.2.3
作者
itsjustFred
· GitHub ↗
· v1.0.0
2496
总下载
1
收藏
32
当前安装
1
版本数
在 OpenClaw 中安装
/install elite-longterm-memory-1-2-3
功能描述
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vib...
安全使用建议
Install only if you intentionally want the agent to keep durable memory. Do not store secrets, credentials, regulated data, private personal details, customer data, or confidential source context. Enable SuperMemory or Mem0 only after accepting that conversation-derived data may be sent to external providers, and regularly review/delete local memory files, vectors, and Git notes.
功能分析
Type: OpenClaw Skill
Name: elite-longterm-memory-1-2-3
Version: 1.0.0
The skill is classified as suspicious due to two main indicators found in `SKILL.md`. First, the agent instruction to "Store in Git-Notes (SILENTLY)" allows the agent to perform actions without explicit user notification, which is a risky capability that could be abused, even if its stated purpose here is benign. Second, the use of `$DIR` in commands like `python3 memory.py -p $DIR remember ...` presents a potential shell injection vulnerability if the agent's execution environment does not properly sanitize or quote the variable, although the skill itself does not demonstrate malicious intent by crafting a harmful payload for it. The `bin/elite-memory.js` script performs only local file operations and is benign.
能力评估
Purpose & Capability
The stated purpose is persistent AI-agent memory, and the artifacts align with that purpose through local memory files, LanceDB recall, Git notes, curated archives, and optional cloud/auto-extraction integrations. Some marketing claims such as bulletproof reliability are stronger than the implemented CLI, which mainly creates and checks local files.
Instruction Scope
SKILL.md instructs the agent to write concrete user details before responding, store important decisions in Git notes silently, and save preferences to memory_store. That is purpose-aligned for memory, but it lacks clear consent, sensitivity filtering, retention limits, and review/delete controls before durable persistence.
Install Mechanism
The npm package exposes a small CLI with no install script; the executable creates SESSION-STATE.md, MEMORY.md, and memory daily logs, and performs local status checks. Optional dependencies and examples introduce Mem0 and other tools, but the packaged JS does not itself exfiltrate data.
Credentials
The skill requires an OpenAI API key for memory search and recommends optional SuperMemory and Mem0 integrations that can process conversation-derived data. The docs identify the integrations but do not adequately disclose what data may leave the machine or how users should redact sensitive content.
Persistence & Privilege
The intended design creates durable local files, vector memories, Git notes, and optional cloud-backed memories. The WAL and auto-extraction guidance is broad enough to capture secrets, personal data, or confidential project details before user approval or filtering.
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 elite-longterm-memory-1-2-3 - 安装完成后，直接呼叫该 Skill 的名称或使用
/elite-longterm-memory-1-2-3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
Elite Longterm Memory 1.2.3 introduces a comprehensive memory system for AI agents with six integrated memory layers.
- Combines HOT RAM, vector search (LanceDB), Git-Notes knowledge graph, curated archives, cloud backup, and auto-extraction (Mem0).
- Ensures persistent and structured context across sessions, with both automated and human-curated long-term memory.
- Adds support for automatic fact extraction and massive token reduction via Mem0.
- Follows a write-ahead log protocol for reliable context retention.
- Includes detailed setup and agent operational instructions for each layer.
元数据
常见问题
Elite Longterm Memory 1.2.3 是什么？
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vib... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 2496 次。
如何安装 Elite Longterm Memory 1.2.3？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install elite-longterm-memory-1-2-3」即可一键安装，无需额外配置。
Elite Longterm Memory 1.2.3 是免费的吗？
是的，Elite Longterm Memory 1.2.3 完全免费（开源免费），可自由下载、安装和使用。
Elite Longterm Memory 1.2.3 支持哪些平台？
Elite Longterm Memory 1.2.3 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 Elite Longterm Memory 1.2.3？
由 itsjustFred（@itsjustfred）开发并维护，当前版本 v1.0.0。
推荐 Skills