← 返回 Skills 市场
michaelfeng

ClawBrain Memory

作者 michaelfeng · GitHub ↗ · v1.2.0 · MIT-0
cross-platform ⚠ suspicious
158
总下载
0
收藏
0
当前安装
8
版本数
在 OpenClaw 中安装
/install clawbrain-pro-memory
功能描述
v1.2 长期记忆系统：自动记住对话内容、区分原话与摘要、长对话不丢失、身份实时更新。让龙虾真正记住一切。
使用说明 (SKILL.md)

ClawBrain Memory Manager

让你的龙虾学会"记住"——不只是当前对话，而是跨对话的结构化长期记忆。

它解决什么问题

OpenClaw 的对话上下文会不断膨胀，最终被截断或压缩，导致：

  • 聊过的事情下次就忘了
  • 用户偏好需要反复说明
  • 重要决策和关键事实丢失
  • 长期运行的 Agent 越用越"笨"

v1.2 核心能力

记忆更可靠

长对话中的内容不再丢失。即使聊了很久，之前提到的关键信息依然会被完整保留，不会因为对话变长而"忘事"。

来源标注

回忆内容时会清晰区分：

  • 你说过的原话：直接引用你的原始表述
  • AI 归纳的摘要：对多次对话内容的智能总结

你可以随时分辨哪些是你亲口说的，哪些是 AI 帮你整理的。

降级透明

当记忆服务暂时不可用时，Agent 会主动告诉你"当前无法访问记忆"，而不是假装什么都不知道或给出错误信息。

身份实时更新

当你告诉 Agent 你换了工作、换了角色、或者任何身份变更，新信息会立即生效，不需要等待或重启。旧的身份信息会被正确归档而非覆盖。

长对话不崩溃

超长对话场景下，系统会智能截断早期低优先级内容，保留重要信息，而不是直接报错或中断对话。

自动整合

每日自动整理记忆：合并重复内容、提炼摘要、清理过时信息，无需手动维护。

使用方法

安装后自动生效。你可以这样触发回忆：

  • "你还记得之前我们讨论过XX吗？"
  • "找一下上次关于XX的对话"
  • "之前那个报告在哪？"

可视化

在 clawbrain.dev/dashboard 的记忆库页面：

  • 记忆浏览和管理
  • 查看记忆来源和时间线

了解更多：https://clawbrain.dev

安全使用建议
Do not install yet. Ask the publisher for concrete implementation details before trusting this skill: where are memories stored (local vs remote), what endpoints are used, what credentials or API keys are required, who can access the stored data, how to view/export/delete memories, encryption/retention policy, and an install/integration guide or source code. Verify the clawbrain.dev domain and the skill author's identity. Prefer skills that publish code or explicit install steps and that declare required env vars and data handling policies before granting access to any sensitive conversations.
能力评估
⚠ Purpose & Capability
The SKILL.md promises persistent, cross‑dialog memory, automatic daily consolidation, real‑time identity updates, and a web dashboard, yet there are no code files, no install steps, no storage/backend details, and no required credentials. A long‑term memory system would normally need storage, an API endpoint, and credentials (or explicit instructions to use built‑in agent storage). The lack of those elements is inconsistent with the claimed capabilities.
⚠ Instruction Scope
The instructions are high‑level marketing/feature descriptions rather than runtime directives. They give no concrete commands, API endpoints, or data handling rules, and they implicitly authorize broad agent behavior (e.g., automatic daily reorganization, identity updates) without specifying limits. This vagueness grants the agent wide discretion and makes it unclear what data will be read, stored, or transmitted.
⚠ Install Mechanism
There is no install spec and no code—while low immediate filesystem risk, it is unexpected for a feature that promises persistent, server‑backed functionality. Normally such a skill would include an install or integration guide (e.g., how to connect to a memory backend, required services). The absence of any implementation detail is a red flag about how the capability would actually be provided.
⚠ Credentials
The skill requests no environment variables or credentials despite describing a dashboard and persistent storage. For a memory service, one would expect API keys, storage configuration, or at least a declaration of where data is kept. This mismatch raises privacy and data‑exfiltration concerns: it's unclear whether memories are local, sent to clawbrain.dev, or handled elsewhere, and there is no mention of access controls or retention policy.
ℹ Persistence & Privilege
always:false and user‑invocable:true (normal). The doc says '安装后自动生效' which implies it activates when installed, but there is no mechanism shown for enabling persistent background tasks. There is no evidence it would modify other skill configs or request elevated system privileges.
如何使用
  1. 确保已安装 OpenClaw（本地或 Docker 部署）
  2. 在对话框中输入安装命令：/install clawbrain-pro-memory
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /clawbrain-pro-memory 触发
  4. 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
v1.2.0
来源标注+降级透明+身份更新+长对话不崩溃
v1.0.2
Fix display name
v1.0.1
v1.0.1: 重写为知识图谱架构 (MySQL + 实体关系 + N-gram 分词 + DreamTask)
v0.9.5
优化 summary 文案
v0.9.4
统一品牌名
v0.9.3
更新 summary 描述
v0.9.2
v0.9.2: 主动思考记忆增强
v1.0.0
- 初始发布：clawbrain-pro-memory 1.0.0 版本上线 - 支持自动记录对话中的关键信息，实现跨对话的长期记忆 - 提供智能回忆机制，可根据历史对话和文件查找上下文 - 支持每日记忆与重要内容沉淀到长期记忆文件 - 安装即用，无需额外配置，提升长期使用的 Agent 记忆力
元数据
Slug clawbrain-pro-memory
版本 1.2.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 8
常见问题

ClawBrain Memory 是什么？

v1.2 长期记忆系统：自动记住对话内容、区分原话与摘要、长对话不丢失、身份实时更新。让龙虾真正记住一切。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 158 次。

如何安装 ClawBrain Memory？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install clawbrain-pro-memory」即可一键安装，无需额外配置。

ClawBrain Memory 是免费的吗？

是的，ClawBrain Memory 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

ClawBrain Memory 支持哪些平台？

ClawBrain 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 ClawBrain Memory？

由 michaelfeng（@michaelfeng）开发并维护，当前版本 v1.2.0。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191113 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论