A long-term memory layer for OpenClaw that enables personalized interactions by persisting user preferences and patterns across sessions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install persistent-user-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 persistent-user-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 Persistent User Memory?
Persistent User Memory transforms your AI from a stateless assistant into a personalized partner. By maintaining a structured local profile, this skill allows Openclaw Skills to remember your specific communication style, professional relationships, and recurring workflows. Instead of repeating instructions every session, the agent consults a dedicated memory store to ensure every action aligns with your established habits and technical requirements.
This skill focuses on making the agent behave like a trusted assistant who actually knows you. It operates by silently reading your user profile before significant actions and updating it as it learns from your feedback and corrections. Because it is stored locally, it provides a high degree of privacy while significantly reducing the cognitive load of re-explaining context to your AI tools.
Persistent User Memory Use Cases
- Maintaining consistent email tones for specific professional contacts without manual prompting.
- Automatically respecting protected calendar hours and preferred meeting buffers during scheduling.
- Remembering technical stacks and preferred coding tools to streamline development tasks.
- Tracking recurring task patterns to automate repetitive administrative duties.
- Recalling past interactions through an episodic log to provide historical context for long-term projects.
How Persistent User Memory Works
- The agent silently loads the local user profile from the file system before performing any significant action.
- Contextual preferences are applied to the current task to ensure high-fidelity results that match user expectations.
- After task completion, the agent identifies new patterns or explicitly stated preferences to update the memory store.
- Significant interactions are logged into a brief episodic history for long-term reference and pattern recognition.
- Automated hygiene processes run periodically to prune old data and keep the memory relevant and accurate.
Persistent User Memory Setup
To add this capability to your environment, use the CLI for Openclaw Skills:
claw skill install persistent-user-memory
Alternatively, you can manually place the skill definition in your local directory:
mkdir -p ~/.openclaw/skills/persistent-user-memory/
# Move the SKILL.md file into this directory
Persistent User Memory Data Schema & Taxonomy
The skill manages a structured JSON profile located at ~/.openclaw/memory/user_profile.json. The data is organized into the following taxonomy:
| Section | Description |
|---|---|
| identity | Basic user info including name, timezone, and language preferences. |
| preferences | Communication styles (tone, length), scheduling rules, and work stacks. |
| relationships | Contact list containing roles, aliases, and specific interaction notes. |
| patterns | Recurring tasks, common mistakes, and frequent user requests. |
| episodic | A chronological log of significant past interactions with summaries and outcomes. |
| meta | Versioning and timestamp data used for automated memory hygiene. |
Persistent User Memory Advanced Features
- Conflict resolution logic that prompts the user for clarification when new instructions contradict stored memory.
- Local-first privacy architecture ensuring that sensitive data is never sent to remote servers.
- Automatic deduplication and stale data flagging during startup routines to maintain data integrity.
- Proactive memory surfacing that warns users of potential errors based on historical failure patterns.
- Surgical data deletion capabilities that allow users to remove specific facts or reset memory entirely.
SKILL.md
Loading
Related Openclaw Skills
WorkCRM
A lightweight, local-first CRM system for AI agents that enforces a strict confirmation gate for all data writes to ensure accuracy and auditability.
Student Timetable
A robust timetable manager supporting both individual student profiles and parent-managed child schedules.
Alibaba Cloud Bailian Model Setup
A streamlined configuration utility for integrating Alibaba Cloud Bailian and Qwen flagship models into your AI workflows.
Daily Briefing Skill
A comprehensive localized dashboard skill that aggregates weather, financial markets, and calendar data into a single visual summary.
Android Remote Control
A technical skill for programmatically controlling Android devices through ADB and uiautomator2 integration.
Gemini Painter
A high-performance AI art generation tool that integrates Gemini Imagen 3 into Openclaw Skills for seamless text-to-image creation.
Featured*
