Dream - Active Memory Distillation for Openclaw

Dream is an automated memory distillation engine that actively maintains your AI agent's core context while archiving historical data into a permanent ledger.

teman2050
v0.2.1
Feb 27, 2026
1
2.3k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install dream-memory-manager

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install dream-memory-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Dream - Active Memory Distillation?

Dream fills a critical gap in native memory management by acting as a proactive distiller rather than a simple capture tool. While standard Openclaw Skills can log interactions, they often suffer from context bloat or silent truncation when files exceed character limits. Dream solves this by monitoring your MEMORY.md file and ensuring it stays under 18,000 characters, keeping your agent's most relevant knowledge within the optimal window for performance.

By implementing a multi-tiered storage strategy, Dream distinguishes between temporary project states, stable cognitive preferences, and historical archives. This ensures that your AI coding agent always has access to high-value context without being overwhelmed by outdated logs or repetitive data.

Dream - Active Memory Distillation Use Cases

  • Automatically distilling daily conversation logs into structured knowledge every night.
  • Preventing context loss by proactively compressing memory files before they hit system limits.
  • Maintaining a permanent, append-only ledger of project decisions and architectural choices.
  • Synchronizing captured knowledge into an Obsidian vault for personal knowledge management.
  • Detecting recurring topics that have re-emerged after being previously archived or forgotten.

How Dream - Active Memory Distillation Works

  1. The skill triggers a daily review at 03:30 or via manual user commands like dream or review.
  2. It reads incremental daily logs and evaluates each entry based on relevance, progress, or obsolescence.
  3. High-value information is used to update the Current State or Stable Cognition sections of the primary memory file.
  4. Completed or aged items are moved to a permanent ledger.md archive and indexed in JSON format.
  5. Before writing, it performs an atomic check to ensure the total character count remains within the safe 18,000-character threshold.
  6. It logs the distillation results and updates the agent's internal state, including re-emergence tracking for recurring concepts.

Dream - Active Memory Distillation Setup

To get started with this component of your Openclaw Skills library, follow these steps:

  1. Ensure you have the required binary dependencies installed:
brew install jq
  1. Configure your environment by setting the vault path:
export DREAM_VAULT_PATH="/your/path/to/vault"
  1. Initialize the memory structure and answer the baseline cognitive questions:
dream init
  1. Add the scheduled trigger to your SOUL.md configuration to enable the 03:30 automated distillation.

Dream - Active Memory Distillation Data Schema & Taxonomy

Dream organizes its data into a structured hierarchy to maximize retrieval efficiency:

File/Directory Purpose Format
MEMORY.md Active context injected into every conversation Markdown (Max 18k chars)
ledger.md Permanent, append-only archive of all historical entries Markdown
ledger-index.json Metadata and IDs for structured searching of the ledger JSON
meta/removed-entries.json Summaries of deleted items used for Re-emergence detection JSON
obsidian-index/ Thematic index files compatible with Obsidian vaults Markdown
meta/dream-state.txt Current operational state (active, hibernating, or pending) Plain Text

Dream - Active Memory Distillation Advanced Features

  • Re-emergence Mechanism: Automatically detects when archived or forgotten topics reappear in conversation and elevates their priority in MEMORY.md.
  • Atomic Write Protection: Uses a temporary file buffer and validation check to prevent data corruption during memory updates.
  • Semantic Search Integration: Wraps native search functionality to provide a unified view across active memory, permanent ledgers, and Obsidian indexes.
  • Hibernation Logic: Automatically enters a zero-I/O state after 7 days of inactivity to save resources, waking up only when a new conversation begins.
  • Dynamic Context Compression: Proactively merges similar entries and archives oldest items first when approaching character limits.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*