A dual-layer AI memory architecture that mimics human biological forgetting and associative learning patterns to optimize context relevance.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-sync-enhanced
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 memory-sync-enhanced 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 Memory Sync Enhanced?
This skill implements a sophisticated memory management system for AI agents by blending two core cognitive theories: the Ebbinghaus Forgetting Curve and Hebbian Co-occurrence. Unlike static vector databases, this Openclaw Skills solution introduces a dynamic decay mechanism that allows agents to prioritize frequently used or recently relevant information while naturally forgetting low-value data.
By layering semantic search with an associative graph, the system ensures that the agent's context window remains high-quality and computationally efficient. It distinguishes between Short-Term Memory (STM) for rapid daily logs and Long-Term Memory (LTM) for permanent insights, creating a more human-like cognitive flow for autonomous agents.
Memory Sync Enhanced Use Cases
- Managing large-scale personal knowledge bases with automated relevance decay.
- Enhancing AI agent personality by establishing cross-domain associative bridges.
- Automating the promotion of daily activity logs into permanent long-term insights.
- Filtering irrelevant noise from agent context windows using biological forgetting logic.
- Optimizing vector database performance through scheduled garbage collection of low-utility memories.
How Memory Sync Enhanced Works
- The system performs an initial semantic search across the CortexGraph vector store to find potential matches.
- A co-occurrence boost is applied via the Hebbian graph to identify associated memories that often appear together in the recent context.
- The Ebbinghaus decay factor calculates the current strength of each memory based on its age, frequency of use, and assigned importance.
- Memories are filtered and ranked based on their final boosted scores to return the most biologically relevant context to the agent.
- A background process handles memory tiering, moving valuable data from STM to LTM or deleting expired nodes.
Memory Sync Enhanced Setup
To integrate this memory system into your Openclaw Skills environment, use the following commands:
# Sync your primary memory database
./scripts/sync-memory.sh
# Synchronize daily logs for a specific date
./scripts/sync-daily.sh 2026-02-19
# Initialize the Hebbian co-occurrence recorder
./scripts/record-co-occurrence.sh
Memory Sync Enhanced Data Schema & Taxonomy
The skill organizes data into three distinct layers to ensure efficient retrieval and persistence:
| Layer | Format | Purpose |
|---|---|---|
| STM | JSONL | Short-term storage with high decay (3-day half-life) |
| LTM | Markdown | Long-term storage for Obsidian-compatible permanent notes |
| Association | SQLite | Tracking co-occurrence weights between memory nodes |
Each memory record includes metadata such as use_count, strength, last_used timestamps, and semantic embeddings.
Memory Sync Enhanced Advanced Features
- Cross-domain bridging to link disparate topics based on simultaneous occurrence.
- Configurable half-life parameters to customize how fast the agent forgets specific types of information.
- Automated garbage collection (GC) to prune memories that fall below a specific relevance threshold.
- Manual promotion workflows to instantly move critical STM entries into LTM storage.
- Detailed memory statistics to visualize the distribution of healthy versus decaying knowledge.
SKILL.md
Loading
Related Openclaw Skills
Memory Sync System (CortexGraph)
An automated memory synchronization system that bridges local Markdown files and CortexGraph using intelligent forgetting curves.
MBTI Agent - Digital Sprite Personality Assessment
A specialized psychometric framework for evaluating the personality profiles of non-interactive AI agents using a static 70-question assessment.
Market Analysis Service
A data-driven analytical suite providing industry trends, competitor deep dives, and user behavior insights.
Lead Research Assistant
An AI-powered assistant designed to identify high-quality sales leads and generate actionable outreach strategies based on your specific product or service.
ML Evolution Agent
An autonomous machine learning agent that iteratively evolves models and leverages multi-layer memory to dominate Kaggle competitions and tabular data tasks.
MLOps Automation
A comprehensive automation suite for standardizing machine learning tasks, containerization, and continuous integration pipelines.