agentsclimarketplace

Persistent memory claw

Skill 2730554131/persistent-memory-claw

OpenCLAW persistent memory skill with multi-agent isolation, vector search, and context inheritance.

Install
npx -y skills add 2730554131/persistent-memory-claw

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

跨会话持久记忆系统:为 OpenClaw agent 提供持久记忆能力。使用 SQLite 存储,按日期保存,手动触发更可控。

SKILL.md

3.7 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

Persistent Memory

为 OpenClaw agent 提供跨会话持久记忆能力。

核心功能

1. 手动保存记忆 (v2.1 智能标记)

当用户说"记住 XXX"、"保存 XXX"时:

const { save } = require('./actions/save');
await save({
  workspace: '{workspace}',
  content: '要记住的内容',
  autoTag: true  // 默认开启智能标记
});

CLI 用法:

# 自动智能标记(默认)
node actions/save.js --content "要记住的内容"

# 指定分类
node actions/save.js --content "任务" --category "task"

# 禁用智能标记
node actions/save.js --content "内容" --no-auto-tag

v2.1 智能标记

  • 自动分析内容重要性(1-10星)
  • 自动分类:task / promise / decision / normal

2. RAG 问答 (v2.2)

当用户问"之前聊过什么?"、"还记得吗?"时:

const { ask } = require('./actions/ask');
await ask({
  workspace: '{workspace}',
  question: '之前我们聊了什么?'
});

CLI 用法:

# 智能问答
node actions/ask.js --question "之前我们聊了什么?"

# 指定日期
node actions/ask.js --question "之前说的密码" --date 2026-03-17

v2.2 RAG 问答

  • 基于记忆的智能问答
  • 自动检索相关记忆
  • LLM 生成自然语言答案

3. 自我反思 (v2.3)

当用户说"反思"、"回顾"、"总结经验"时:

const { reflect } = require('./actions/reflect');
await reflect({
  workspace: '{workspace}',
  days: 7
});

CLI 用法:

# 自我反思(最近7天)
node actions/reflect.js

# 回顾最近30天
node actions/reflect.js --days 30

v2.3 自我反思

  • 定期回顾近期记忆
  • 提取经验教训
  • 总结主要工作
  • 识别待完成任务

4. 生成摘要

const { summarize } = require('./actions/summarize');
await summarize({
  workspace: '{workspace}',
  date: '2026-03-17'
});

CLI 用法:

node actions/summarize.js
node actions/summarize.js --date 2026-03-17

3. 搜索记忆

当用户说"搜索 XXX"、"查找 XXX"时:

const { search } = require('./actions/search');
await search({
  workspace: '{workspace}',
  query: '关键词',
  date: '2026-03-17'
});

CLI 用法:

node actions/search.js --query "关键词"
node actions/search.js --query "关键词" --date 2026-03-17

4. 列出记忆

当用户说"查看记忆"、"列出记忆"时:

const { list } = require('./actions/list');
await list({
  workspace: '{workspace}',
  date: '2026-03-17'
});

CLI 用法:

node actions/list.js
node actions/list.js --date 2026-03-17

存储结构

{workspace}/memory/
├── 2026-03-17.db     # 3月17日的记忆
├── 2026-03-16.db     # 3月16日的记忆
├── 2026-03-17-summary.md  # 摘要文件
└── ...

SQLite 表结构

CREATE TABLE memories (
  id INTEGER PRIMARY KEY AUTOINCREMENT,
  content TEXT NOT NULL,
  category TEXT DEFAULT 'default',
  created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

意图识别

功能触发关键词
保存记忆(智能标记)记住、保存、记录、存一下
RAG 问答之前、记得、聊过、问
自我反思反思、回顾、总结经验
生成摘要摘要、总结、提炼
搜索记忆搜索、查找、找一下、记得
列出记忆查看记忆、列出记忆、今天的记忆

依赖

  • Node.js
  • sqlite3(首次使用自动安装)

Gives 0 of the 12 instructions most memory context skills give in ~1.2k tokens

Counted across 674 of the 847 authors here whose files we hold, read 2026-08-06

  • inform the user when setup is completein 21 of 674, across 6 files
  • confirm the draft with the user before writingin 21 of 674, across 6 files
  • update the agent skills block in place if it existsin 21 of 674, across 6 files
  • present findings to the userin 20 of 674, across 5 files
  • write the three docs files from seed templatesin 20 of 674, across 5 files
  • ask the user about each decision one at a timein 19 of 674, across 4 files
  • edit CLAUDE.md if it existsin 18 of 674, across 3 files
  • explore current repo statein 18 of 674, across 3 files
  • do not overwrite user edits to surrounding sectionsin 18 of 674, across 3 files
  • back up the original file before overwritingin 16 of 674, across 8 files
  • keep the memory index under 200 linesin 15 of 674
  • Provide actionable steps and verificationin 13 of 674, across 2 files

Said here and by no other author read

  • use SQLite to store memories by date
  • save memory with auto-tagging enabled
  • answer questions using retrieval-augmented generation
  • generate reflections on recent memories
  • generate summaries by date
  • list memories by date

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.