A local, zero-cost persistent memory system that enables AI agents to remember user preferences and project context without external APIs.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openjaw-smart-memory
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 openjaw-smart-memory 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 Smart Memory?
Smart Memory provides a robust, file-based intelligence layer for your AI agents. By utilizing local JSON storage, it allows agents to extract, store, and retrieve key facts, technical decisions, and user preferences across multiple sessions. This ensures a personalized experience where the agent learns over time without incurring additional API costs or privacy risks.
The system is designed to live entirely on your local machine within the Openclaw Skills ecosystem, keeping your data private and secure. It automates the process of context retention, making every interaction more efficient by reducing the need for repetitive instructions and significantly lowering token usage over time.
Smart Memory Use Cases
- Remembering developer preferences like coding styles, preferred libraries, or IDE themes.
- Storing project-specific decisions such as database choices, API endpoints, or architectural patterns.
- Tracking personal facts and roles within a team to personalize communication.
- Logging technical context like repository URLs and server addresses for quick retrieval.
- Persisting instructions that should apply to all future tasks without re-prompting.
How Smart Memory Works
- The agent searches for existing memories using keywords before generating a response to ensure context awareness.
- When new information is shared (preferences, facts, or decisions), the agent invokes the store command to save it locally.
- Memories are assigned confidence levels and source types to ensure the reliability of retrieved data.
- The agent updates existing entries when corrections are provided, maintaining a historical audit trail.
- A daily heartbeat trigger performs maintenance tasks like pruning stale data and deduplicating entries.
Smart Memory Setup
To integrate Smart Memory into your Openclaw Skills environment, ensure the scripts are placed in the correct directory and are executable.
# Create the directory
mkdir -p ~/.openclaw/smart-memory/
# Move the memory manager scripts to the folder
mv memory-manager.sh memory-stats.sh ~/.openclaw/smart-memory/
# Make them executable
chmod +x ~/.openclaw/smart-memory/*.sh
Smart Memory Data Schema & Taxonomy
The skill organizes data within ~/.openclaw/smart-memory/ using a structured JSON approach to manage Openclaw Skills data:
| File | Description |
|---|---|
memories.json |
The main store for active memories and context. |
archive.json |
Stores pruned or soft-deleted memories for a 30-day retention period. |
stats.json |
Tracks usage metrics and estimated token savings. |
config.json |
Contains user-defined settings for maintenance intervals and confidence thresholds. |
Smart Memory Advanced Features
- Automatic Token Savings Tracking to calculate the ROI of persistent context.
- Multi-level confidence scoring (high, medium, low) for nuanced data retrieval.
- Daily automated heartbeat for self-healing, deduplication, and integrity checks.
- Privacy-first local-only storage ensuring no sensitive data leaves the machine.
- Soft-delete functionality with a 30-day recovery window for forgotten information.
SKILL.md
Loading
Related Openclaw Skills
Bot Arcade
An all-in-one text-based gaming and entertainment engine that turns AI agents into interactive arcade hosts.
ConvoYield
ConvoYield is a real-time revenue intelligence layer that transforms every bot interaction into a yield-bearing financial instrument through behavioral economics.
Negative Space
A metacognitive reasoning skill that trains AI agents to identify missing context and hidden assumptions before executing any task.
Naming Forge
A sophisticated naming engine that uses linguistic principles and codebase analysis to generate precise, consistent, and discoverable names for any programming element.
ResonanceEngine
ResonanceEngine is a zero-cost algorithmic tool that analyzes conversational micro-signals to optimize AI bot responses for maximum engagement and conversion.
Osmotic Pressure
A sophisticated diagnostic tool that identifies and maps complexity imbalances across software boundaries to prevent architectural collapse.
Featured*
