A Retrieval-Augmented Generation (RAG) based intelligence system providing semantic search and memory optimization for Openclaw Skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install smart-memory-system
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 smart-memory-system 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 System?
The Smart Memory System is a high-performance framework designed to provide persistent, context-aware intelligence for Openclaw Skills. By leveraging Retrieval-Augmented Generation (RAG) technology, it allows AI agents to move beyond simple keyword matching to true semantic understanding. This skill ensures that your agent retains a long-term memory of user preferences, project histories, and technical documentation across multiple sessions.
Built for efficiency, this system significantly optimizes resource usage by reducing token consumption by up to 80%. It employs sophisticated embedding models and re-ranking algorithms to ensure that only the most relevant information is injected into the AI's context window, transforming a standard coding assistant into a deeply personalized and highly accurate knowledge powerhouse.
Smart Memory System Use Cases
- Personal Assistant Enhancement: Maintain user habits, cross-session continuity, and personalized suggestion generation.
- Team Knowledge Management: Enable shared knowledge base retrieval and historical project tracking for collaborative environments.
- Research and Analytics: Perform intelligent literature retrieval and organize research notes into actionable insights.
- Developer Support: Conduct semantic searches across codebases and match errors with historical solutions instantly.
How Smart Memory System Works
- Vectorization: The system uses the BAAI/bge-m3 embedding model to convert text inputs into 1024-dimensional vectors.
- Semantic Storage: Memories are stored in a local JSON structure with an integrated semantic cache for ultra-fast retrieval.
- Retrieval and Re-ranking: Upon a query, the system identifies relevant chunks using cosine similarity and refines them via the bge-reranker-v2-m3 module.
- Importance Scoring: Memories are evaluated based on age, frequency of use, and relevance to ensure priority data is prioritized.
- Automated Injection: Relevant historical context is automatically injected into the active Openclaw Skills conversation window.
Smart Memory System Setup
To get started with this memory system for Openclaw Skills, ensure you have Node.js and an Edgefn API key for the embedding models. Follow these steps:
# Install via ClawHub
clawhub install smart-memory-system
# Initialize the memory system
openclaw skill smart-memory init
# Load existing documents into memory
openclaw skill smart-memory batch-process ~/documents/
You must also configure your provider in the OpenClaw settings to support BAAI/bge-m3 and bge-reranker-v2-m3 models.
Smart Memory System Data Schema & Taxonomy
The system organizes data using a structured hierarchy to ensure rapid retrieval and metadata integrity. Below is the primary file taxonomy:
| Component | Description |
|---|---|
| smart_memory.json | Main configuration for chunk size, overlap, and similarity thresholds. |
| models.json | Defines the embedding and re-ranking model providers. |
| vectorizer.js | Core logic for transforming text into 1024-dim vectors. |
| retriever.js | Engine for semantic search and re-ranking logic. |
| integrator.js | The bridge for seamless Openclaw Skills integration. |
Smart Memory System Advanced Features
- Batch processing of local directories for rapid knowledge base ingestion.
- Aggressive index optimization to maintain high performance as memory grows.
- Customizable importance scoring with adjustable weights for age, frequency, and relevance.
- Multi-format data export supporting JSON, Markdown, CSV, and HTML reports.
- Plugin architecture allowing developers to register custom memory processors and templates.
SKILL.md
Loading
Related Openclaw Skills
Memory Core
An intelligent, modular memory system for AI agents that integrates semantic search, multi-provider embeddings, and Flomo notes.
Superdesign
A specialized design agent that automates frontend UI/UX workflows, from initial repo analysis to component extraction.
My Generate QR Code
An AI-powered utility for generating highly customizable QR codes and barcodes for various data types including URLs and WiFi configurations.
Folder Inspector
A high-performance AI agent skill for scanning local directories and auditing file metadata using absolute paths.
Humanizer
A sophisticated writing editor that identifies and removes common AI-generated signatures to create natural, human-sounding text.
Fliz AI Video Generator
Programmatically transform text content into professional AI-generated videos with voiceovers and automated subtitles.
Featured*
