A lightweight, file-based memory system that allows AI agents to persist learnings and decisions with sub-20ms search performance.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-system-v2
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-system-v2 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 System v2.0?
Memory System v2.0 is a specialized solution for AI agents that struggle with context loss between sessions. Built using pure bash and jq, it provides a high-performance, database-free architecture for storing and retrieving semantic memories. This tool is a critical addition to the library of Openclaw Skills, enabling agents to track learnings, decisions, insights, and interactions with millisecond precision.
By utilizing a JSON-indexed file structure, the system ensures that agents can recall prior work and project history without the overhead of external databases. It bridges the gap between ephemeral chat sessions and long-term knowledge retention, making it easier for developers to build sophisticated agents that grow more capable over time through continuous context accumulation.
Memory System v2.0 Use Cases
- Persistent tracking of technical skills and coding patterns learned during development sessions.
- Maintaining a historical audit trail of architectural decisions and strategic pivots.
- Automatically generating weekly progress reports and executive summaries from daily activity logs.
- Improving user experience by recalling specific feedback and preferences from previous interactions.
- Enabling agents to search through large project histories to find relevant past milestones or events.
How Memory System v2.0 Works
- The agent captures a memory using the CLI, assigning it a specific type and importance score between 1 and 10.
- The system appends the full memory details to a daily Markdown log for human-readable persistence.
- Simultaneously, the system updates a centralized JSON index to facilitate rapid semantic search and filtering.
- When a search is triggered, the jq-powered engine scans the index to return relevant memories in less than 20ms.
- An automated consolidation process periodically sweeps daily logs to create structured weekly summaries for high-level review.
Memory System v2.0 Setup
To integrate this tool with your Openclaw Skills workflow, follow these steps:
# Install the required jq dependency for JSON processing
brew install jq
# Ensure the memory-cli.sh script is executable within your workspace
chmod +x ./memory/memory-cli.sh
# Initialize your first memory capture
./memory/memory-cli.sh capture --type learning --importance 9 --content "Initial setup complete" --tags "setup,init"
Memory System v2.0 Data Schema & Taxonomy
The system organizes data into a tiered directory structure for maximum efficiency and readability:
| File/Folder | Purpose |
|---|---|
memory/index/memory-index.json |
High-speed search index containing metadata and file references. |
memory/daily/YYYY-MM-DD.md |
Chronological logs of every memory captured on a specific date. |
memory/consolidated/YYYY-WW.md |
Weekly summaries generated by the auto-consolidation engine. |
Each memory object in the JSON index includes fields for unique ID, importance rating, timestamp, content body, context strings, and an array of tags.
Memory System v2.0 Advanced Features
- Smart multi-word search functionality with importance-weighted filtering for accurate recall.
- Automated weekly consolidation scripts that synthesize daily logs into high-level summaries.
- Built-in analytics engine to track memory counts, type distribution, and priority trends.
- Native integration with agent protocols via AGENTS.md for autonomous memory retrieval.
- Optimized performance profile ensuring all CLI operations execute in under 100ms.
SKILL.md
Loading
Related Openclaw Skills
Agentic Calling Skill
A comprehensive toolkit that enables AI agents to autonomously handle outbound and inbound phone calls using Twilio integration.
brave-headless
A sophisticated headless search and web scraping tool that leverages the Brave Search API to provide structured, markdown-formatted web content.
Document Summary
An automated technical analyst that transforms complex text into structured markdown summaries, key points, and risk assessments.
Document Summary
A professional AI-driven skill that converts long technical documents into structured Markdown summaries including key points and risks.
OpenClaw Deck
A powerful multi-column web interface for side-by-side management and interaction with multiple OpenClaw agents.
Portfolio Risk & Optimization Analyzer
An AI-powered crypto portfolio analyzer that delivers deep risk metrics, stress testing, and automated optimization strategies while supporting a token buyback ecosystem.
Featured*
