Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Updated skill description, fixed opening paragraph
What This Skill Does
Saves and retrieves agent memory using semantic search, so context persists across sessions. Manual commands let you store and query specific text; optional hooks automate capture and recall around each AI turn.
Semantic search finds relevant context by meaning rather than requiring exact keyword matches, which makes recall reliable even when you don't remember how something was phrased.
When to use it
- Remembering user preferences set in a previous session
- Recalling a decision made days ago without re-explaining it
- Automatically injecting relevant past context before each AI response
- Storing project-specific conventions the agent should always follow
- Searching past conversations by meaning rather than exact keywords
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'What font size did we decide on for the dashboard?'
- 1Auto-Recall hook triggers before the AI turn
- 2Sends the user message as a search query to POST /memory/search
- 3Neutron API returns semantically matched memories about past font size decisions
- 4Relevant context is injected into the AI turn as additional context
- 5Agent responds with the recalled decision
Agent answers with the font size choice from a previous session, without the user repeating themselves
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.