A quantum-optimized memory framework that uses QAOA to solve clustering, compaction, and synergy-based recall problems for AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install quantum-agent-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 quantum-agent-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 Quantum Agent Memory?
Quantum Agent Memory is a high-performance optimization layer built to enhance the memory management capabilities of autonomous agents. By utilizing the Quantum Approximate Optimization Algorithm (QAOA) through Qiskit, this tool provides a robust alternative to classical heuristics for managing complex data relationships. As a sophisticated addition to Openclaw Skills, it enables developers to implement quantum-enhanced clustering, compaction, and recall strategies that significantly improve context retention.
The skill addresses the limitations of traditional vector search by considering the synergy between retrieved memories rather than just individual similarity scores. This ensures that agents powered by Openclaw Skills can maintain a more coherent and relevant long-term memory, whether operating on local simulators or actual IBM Quantum hardware processors.
Quantum Agent Memory Use Cases
- Grouping large sets of memories into logical clusters using balanced graph-cut optimization.
- Performing intelligent memory compaction to retain the most valuable context within token constraints.
- Implementing synergy-aware memory recall that outperforms standard Top-K similarity searches.
- Benchmarking the accuracy of quantum memory selection against classical algorithms within Openclaw Skills pipelines.
- Automating the submission of quantum circuits to IBM Quantum hardware for real-world agent memory tasks.
How Quantum Agent Memory Works
- The system ingests memory data and generates a cost matrix based on temporal, relational, categorical, and recency factors.
- The Clustering Layer applies QAOA to group memories into coherent clusters based on the cost matrix.
- The Compaction Layer selects the optimal K-memories from a larger set to maximize coherence and coverage while penalizing budget overages.
- The Recall Layer identifies the best combination of memories for a specific query by optimizing for individual relevance and pairwise synergy.
- The processed results are delivered via a FastAPI server or integrated into an agent workflow via standard Openclaw Skills interfaces.
Quantum Agent Memory Setup
To integrate this powerful tool from the Openclaw Skills collection into your environment, follow these installation steps:
git clone https://github.com/Dustin-a11y/quantum-agent-memory.git
cd quantum-agent-memory
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
python -m quantum_agent_memory benchmark
For users intending to utilize real IBM Quantum hardware, additional setup is required:
pip install qiskit-ibm-runtime
python -m quantum_agent_memory submit --ibm-token YOUR_TOKEN
Quantum Agent Memory Data Schema & Taxonomy
The skill manages data through a multi-layered schema designed for deep integration with Openclaw Skills workflows:
| Component | Purpose | Key Metadata |
|---|---|---|
| Memory Clusters | Organizing related data points | Recency, category weight, relational strength |
| Compaction Sets | Pruned memory for token efficiency | Coherence score, value density, coverage percentage |
| Recall Synergy | Multi-memory retrieval logic | Pairwise synergy, diversity index, relevance |
| Performance Logs | Comparative analysis | Quantum vs classical delta, circuit depth, execution time |
All output data is exported as structured JSON files within the results/ directory for easy parsing.
Quantum Agent Memory Advanced Features
- Full compatibility with Mem0 for managing live agent memory streams.
- FastAPI server implementation for remote quantum optimization via the quantum-recall and quantum-compact endpoints.
- Automated scheduling for IBM Quantum hardware jobs via specialized cron scripts.
- Native support for the OpenClaw mem0-bridge plugin, facilitating seamless Openclaw Skills integration.
- Detailed benchmarking suite providing statistical proof of quantum optimization advantages over greedy selection methods.
SKILL.md
Loading
Related Openclaw Skills
X Ads CLI
A read-only command-line interface for querying X Ads API v12 to analyze campaign performance and account structure.
Executive Chief of Staff
An AI-powered chief of staff designed to optimize executive decision-making through automated briefings, meeting prep, and strategic information gathering.
Healthcare & Clinic Operations
An AI-powered operational assistant designed to manage patient scheduling, medical records, and clinic workflows for healthcare facilities.
Government & Public Sector Skill
A comprehensive automation skill for government ministries and public institutions to manage formal documentation, reporting, and inter-departmental communication.
Feishu Notifier Core
A lightweight, secure bridge for sending automated notifications to Feishu (Lark) through Openclaw Skills.
Spotify Ads CLI
A read-only CLI tool for the Spotify Ads API (v3) that enables deep reporting, audience estimation, and campaign analysis.
Featured*
