ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified productivity Safety 4/5

dream-cycle

Implement a nightly "dream cycle" for OpenClaw agents that: - Reviews and consolidates memory at night - Cleans up bloated workspace files - Prepares a morning brief for the user Use when: User wants to implement Ray Fernando's dream cycle concept, reduce memory bloat, or set up automated memory maintenance.

Why use this skill?

Implement the dream-cycle skill to keep your OpenClaw agent lean and organized. Automatically prune memory bloat, synthesize project history, and get daily briefs.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/imjohnathanblog-spec/dream-cycle
Or

What This Skill Does

The dream-cycle skill is an automated, high-level memory maintenance system for your OpenClaw agent, inspired by Ray Fernando’s concept of cognitive maintenance. It functions as a nightly cleanup routine that mimics human sleep patterns—consolidating long-term memories, pruning redundant information, and preparing the agent for the upcoming day. By systematically managing your Tier 1 and Tier 2 memory files (such as AGENTS.md and MEMORY.md), the skill ensures that your agent remains sharp, lightweight, and focused on current goals without being hindered by context bloat.

Installation

To integrate this into your workflow, use the terminal command: clawhub install openclaw/skills/skills/imjohnathanblog-spec/dream-cycle. Once installed, you will need to configure your cron schedule. The system requires two primary cron jobs: one at 03:00 to perform the memory audit and pruning (the 'Dream' phase) and one at 07:00 to generate and deliver your morning summary (the 'Brief' phase). Ensure the agent has read/write permissions for the memory directories to allow for automatic file optimization.

Use Cases

This skill is ideal for power users who rely on their OpenClaw agent for long-term project tracking. It is best suited for scenarios where:

  • You notice your agent is becoming 'forgetful' or losing track of context due to file size limits.
  • You want to maintain a clean workspace by purging outdated temporary logs.
  • You need a structured morning routine that summarizes progress from the previous day.
  • You are managing complex, multi-week tasks that require periodic historical synthesis to keep the agent's reasoning grounded.

Example Prompts

  1. "OpenClaw, execute a dream cycle right now to clear out the bloat from our recent research files."
  2. "Can you perform a dream audit and show me which sections of MEMORY.md are currently taking up the most space?"
  3. "Please run the dream brief command so I can see a summary of what we accomplished yesterday before I start my work."

Tips & Limitations

To maximize the efficiency of the dream-cycle, keep your core files (AGENTS.md) under 2,000 bytes. The skill relies heavily on QMD (Queryable Memory Data) for indexing, so ensure your memory storage is properly configured for indexing tasks. Note that during the 'Dream' phase, the agent may modify files; always keep backups if you are working on highly sensitive data. Additionally, ensure your system clock is correctly synchronized, as the cron-based automation relies on precise timing to execute correctly at 3 AM and 7 AM.

Metadata

Stars2287
Views1
Updated2026-03-09
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-imjohnathanblog-spec-dream-cycle": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#memory-management#automation#productivity#optimization
Safety Score: 4/5

Flags: file-read, file-write