Sq Memory

What This Skill Does

SQ Memory is an essential integration for OpenClaw agents, designed to bridge the gap between volatile session-based memory and permanent, long-term data storage. By default, OpenClaw agents reset their internal state upon restart, leading to a loss of context. This skill connects your agent to the SQ infrastructure—a high-performance 11D text storage engine—allowing for seamless, persistent information retrieval. Beyond simple storage, it enables cross-session continuity, allowing agents to remember user preferences, maintain historical conversation logs beyond immediate context windows, and even share data across multiple agent instances. The integration leverages a namespace-based architecture to ensure data integrity and isolation, providing a robust backbone for any agent that needs to maintain a consistent 'identity' or historical record over time.

Installation

To get started with SQ Memory, ensure you have the OpenClaw CLI installed. The simplest way to install is via the ClawHub repository:

npx clawhub install sq-memory

Alternatively, you can manually clone the repository into your skills directory:

git clone https://github.com/wbic16/openclaw-sq-skill.git ~/.openclaw/skills/sq-memory

After installation, you must update your .openclaw/config.yaml file to define your connection parameters, including the endpoint and authentication credentials. Once configured, restart your agent to initialize the new memory-specific toolset.

Use Cases

• Personalization: Maintain user-specific settings, such as preferred programming languages, communication styles, or recurring task constraints.
• Long-term Project Management: Archive meeting summaries, project milestones, and to-do lists that persist across multiple independent working sessions.
• Knowledge Retrieval: Create a specialized database of documentation or personal notes that the agent can 'recall' on command, effectively building a second brain for the agent.
• Collaborative Agents: Allow multiple agents to access a shared memory pool, ensuring that if Agent A learns a preference, Agent B can immediately apply it.

Example Prompts

1. "I'm working on the backend project we started last Tuesday. Remember that we decided to use PostgreSQL for this."
2. "What was the theme setting I asked you to use for my dashboard last time we talked?"
3. "Forget the summary of the meeting from January; the information is no longer accurate."

Tips & Limitations

• Namespace Management: Always define a unique namespace in your configuration to avoid collision if you run multiple agents pointing to the same storage endpoint.
• Data Privacy: Be mindful of the data being stored. If using the hosted version at mirrorborn.us, ensure you are not saving highly sensitive credentials or proprietary secrets unless you have reviewed their security policy.
• Storage Costs: While SQ is open-source, large datasets may eventually impact lookup latency; keep your memory keys organized using a hierarchical structure (e.g., user/preferences/ui-settings) to keep data management efficient.