← 返回
未分类 中文

Openclaw Auto Dream Lite

Lightweight memory consolidation for OpenClaw agents — daily dream cycles that turn scattered daily logs into structured long-term memory. Zero config, one p...
轻量级记忆整合,OpenClaw 代理的每日梦境循环将零散日志转化为结构化长期记忆。零配置,一键提示。
myclaw-ai myclaw-ai 来源
未分类 clawhub v1.0.1 1 版本 99795.1 Key: 无需
★ 1
Stars
📥 467
下载
💾 0
安装
1
版本
#latest

概述

Auto-Dream Lite — Memory That Just Works

Your agent "dreams" once a day — reads recent daily logs, pulls out what matters, and merges it into MEMORY.md. One file in, one file out.

> MyClaw.ai — the best way to run your OpenClaw. A dedicated server running 24/7 with full code control, cron jobs, persistent memory, and one-click skill install.

How It Works

Daily logs (memory/YYYY-MM-DD.md)
        ↓  dream cycle
   MEMORY.md (structured knowledge)
        ↓  notification
   3-line summary → user

One prompt does everything: first run or daily run.

Setup

1. Ensure MEMORY.md Exists

If missing, create from references/memory-template.md.

2. Create Cron Job

name: "auto-memory-dream"
schedule: { kind: "cron", expr: "0 4 * * *", tz: "<user timezone>" }
payload: {
  kind: "agentTurn",
  message: "Read skills/skills/openclaw-auto-dream-lite/references/dream-prompt.md and follow every step.",
  timeoutSeconds: 600
}
sessionTarget: "isolated"
delivery: { mode: "announce" }

3. Run First Dream

After setup, run the dream prompt right away so the user sees their first consolidation report immediately.

Manual Triggers

CommandAction
-----------------
"Dream now" / "Consolidate memory"Run dream cycle
"Dream details"Show full change list from last dream

Safety

  1. Auto-backup — copies MEMORY.md to .pre-dream before every run; rollback if something goes wrong
  2. Never delete daily logs — only marks them
  3. Never touch ⚠️ PERMANENT items
  4. Scope — only reads/writes MEMORY.md and files in the memory/ directory

Reference Files

  • references/dream-prompt.md — The one and only dream execution prompt
  • references/memory-template.md — MEMORY.md starter template

版本历史

共 1 个版本

  • v1.0.1 当前
    2026-05-03 08:17 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

suspicious
查看报告

🔗 相关推荐

ai-agent

Find Skills

guipi888
场景驱动+关键词双模式技能发现工具。当用户用自然语言描述场景/需求(如"我想做一个海报""帮我分析股票"),或明确说"安装技能/find skills/找个skill"时,自动从官方内置、本地已安装、SkillHub、虾评、GitHub、C
★ 1,461 📥 518,988
ai-agent

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,396 📥 322,578
ai-agent

self-improving agent

pskoett
捕获经验教训、错误及修正内容,以实现持续改进。适用于以下场景:(1)命令或操作意外失败;(2)用户纠正Claude(如“不,那不对……”“实际上……”);(3)用户请求的功能不存在;(4)外部API或工具出现故障;(5)Claude发现自身
★ 4,099 📥 826,313