Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Replaced LICENSE.md with LICENSE for standard naming. - Updated installation instructions in documentation to use the new `npx clawhub install sq-memory` method. - Configuration examples now recommend using `http://localhost:1337` as the default endpoint. - Adjusted self-hosted and hosted setup steps in the documentation for clarity. - Minor edits for documentation consistency and updated tool/feature naming.
What This Skill Does
Connects OpenClaw agents to SQ, a persistent 11D text storage system. Agents can store and retrieve named memories across sessions using coordinate-based keys. Supports multi-agent memory sharing through shared coordinates.
Unlike vector databases or Redis, SQ uses deterministic coordinate-based retrieval so agents can read exact stored values without similarity ranking or data loss on restart.
When to use it
- Remembering user preferences between chat sessions
- Storing conversation summaries past context window limits
- Coordinating task handoffs between multiple agents
- Persisting user profile data without re-asking each session
- Sharing state between specialized agents in a pipeline
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: "I prefer dark mode" in one session, then later asks "What's my preferred theme?" in a new session
- 1Agent calls remember("user/preferences/theme", "dark") to store the preference
- 2Memory is written to SQ at a namespaced 11D coordinate
- 3New session starts; agent has no in-context memory of prior conversation
- 4Agent calls recall("user/preferences/theme") to retrieve the stored value
- 5Agent returns "You prefer dark mode" based on persisted data
Agent correctly answers with the stored preference despite session restart
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.