A sophisticated memory maintenance tool that distills daily logs into thematic topic files while pruning outdated information.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dream-rem
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install dream-rem 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 Rem?
Dream Rem acts as a cognitive processing layer for AI agents, functioning similarly to a dream cycle by consolidating short-term daily logs into long-term thematic memory. By utilizing these Openclaw Skills, your agent can prevent memory bloat and maintain a high-performance context window. The system analyzes the last 14 days of activity, extracts significant insights, and organizes them into a structured directory of topic-specific markdown files.
This skill is essential for developers who need their agents to maintain long-term coherence without exceeding token limits. It transforms the primary MEMORY.md file into a lean, high-speed index that points to detailed records, ensuring the agent always has access to the most relevant and up-to-date information while discarding contradictions and obsolete data.
Dream Rem Use Cases
- Consolidating fragmented daily journals into unified thematic topic files.
- Automatically pruning contradictory or outdated information to maintain a single source of truth.
- Managing agent memory size to prevent performance degradation over long-running sessions.
- Converting relative time references in logs into absolute dates for better historical tracking.
How Dream Rem Works
- The skill triggers automatically via cron or manual command, checking session count and time elapsed since the last consolidation.
- It scans the topics directory and the existing memory index to establish a baseline of known information.
- A strict 14-day window of daily log files is gathered and analyzed, ensuring no files are skipped.
- New information is identified and either appended to existing topics or used to generate new ones.
- Outdated pointers and contradictory versions are pruned, and the MEMORY.md index is rewritten to stay within size limits.
- An execution report is generated, detailing exactly how many memories were consolidated or cleared.
Dream Rem Setup
To install and configure Dream Rem, initialize your state file and set up a recurring cron job. First, ensure memory/heartbeat-state.json exists with initial values: {"lastExtraction": null, "lastDreamAt": null, "sessionCount": 0}. Then, execute the following command to enable automated memory maintenance:
openclaw cron add --name "Memory Deep Consolidation" --every 12h --session isolated --timeout-seconds 600 --message "Check and execute dream-rem memory consolidation..." --announce
Dream Rem Data Schema & Taxonomy
Dream Rem organizes data through a hierarchical structure focused on retrieval efficiency:
| Component | Location | Role |
|---|---|---|
| Index File | MEMORY.md |
A compact list of pointers (max 200 lines) linking to topics. |
| Topic Files | topics/*.md |
Thematic files containing distilled, long-term memory. |
| Daily Logs | memory/*.md |
Raw daily activity logs used as the primary data source. |
| State Tracker | heartbeat-state.json |
Metadata tracking session counts and last execution timestamps. |
Dream Rem Advanced Features
- Conflict resolution logic that identifies and removes contradictory information rather than duplicating it.
- Strict file-scanning gates that prevent progress unless all files in the 14-day window are successfully processed.
- Automatic relative-to-absolute date normalization for all distilled memory entries.
- Metadata-driven triggers that prevent unnecessary processing until specific session or time thresholds are met.
SKILL.md
Loading
Related Openclaw Skills
Knab
A specialized on-chain research agent for the AIMS protocol that provides deep visibility into vault yields, reserves, and transaction data across multiple blockchains.
OpenClaw Trust Validation
A robust validation framework that prevents OpenClaw agents from blindly trusting outdated or incorrect memory during decision-making.
Memory-Sorting
A systematic maintenance skill that organizes, prunes, and validates AI agent memory files to prevent information decay.
WeChat MP Reader
A production-ready skill for AI agents to directly extract titles and full body text from WeChat Official Account links using Playwright.
OpenClaw Memory Freshness
A sophisticated utility for OpenClaw that injects time-based freshness warnings into memory search results to ensure AI agents validate outdated information.
OpenClaw Compact Improver
A professional optimization tool for the OpenClaw /compact command that implements a 9-part structured summary to ensure high-fidelity context retention.
Featured*
