A production-grade integration that provides AI agents with persistent long-term memory capabilities via the EverMemOS system.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install evermemos
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 evermemos 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 EverMemOS Memory System?
The EverMemOS skill is a sophisticated memory integration designed to give AI agents a persistent brain. By leveraging Openclaw Skills, this integration allows agents to store and recall information across different sessions, moving beyond the limitations of standard context windows. It acts as a specialized database for memories, categorizing information into episodic events, user profiles, and future plans.
This skill is essential for developers looking to create personalized AI assistants that grow with the user. It supports multiple retrieval methods, including keyword, semantic vector, and hybrid searches, ensuring that the agent can always find the most relevant context to inform its current task.
EverMemOS Memory System Use Cases
- Persistent storage of user preferences and profile information for a personalized experience.
- Semantic retrieval of past conversations to maintain context across multi-day sessions.
- Automated extraction and storage of key takeaways or action items after a task is completed.
- Recalling specific technical details or past deployment steps during development workflows.
- Tracking and updating long-term user goals and future plans for proactive assistance.
How EverMemOS Memory System Works
- The AI agent identifies a trigger (e.g., 'remember this' or a conversation end) and extracts key information.
- The data is categorized into specific memory types such as Episodic Memory or User Profiles.
- The skill sends the processed content to the EverMemOS server via a secure API request.
- During retrieval, the agent executes a search query (keyword, vector, or hybrid) to find relevant historical data.
- The retrieved memories are returned to the agent's prompt, providing it with deep historical context.
EverMemOS Memory System Setup
To utilize this skill, you must have an active EverMemOS server instance.
- Deploy EverMemOS Server:
# Clone the repository and start via Docker
cd EverMemOS
docker-compose up -d
# Verify the service is running
curl http://localhost:1995/health
- Configure Environment Variables: Set the following variables in your environment to connect the skill:
EVERMEMOS_URL: The API address of your server (Default: http://localhost:1995).EVERMEMOS_API_KEY: Your authentication key if applicable.
- Test Memory Storage:
curl -X POST ${EVERMEMOS_URL}/api/v1/memories \
-H "Content-Type: application/json" \
-d '{"content": "User prefers Python for data science", "sender": "user", "user_id": "user_01"}'
EverMemOS Memory System Data Schema & Taxonomy
The skill organizes memory into a structured taxonomy for optimized retrieval:
| Memory Type | Purpose | Example |
|---|---|---|
| EPISODIC_MEMORY | Conversation events | "User learned AI deployment today" |
| PROFILE | User personas | "User prefers minimalist UI designs" |
| FORESIGHT | Future planning | "User intends to learn LangChain next month" |
| EVENT_LOG | Atomic facts | "User deployed MongoDB successfully" |
Each record includes a message_id, content, sender, timestamp, and scene metadata.
EverMemOS Memory System Advanced Features
- Hybrid Search: Combines keyword precision with semantic vector understanding for the most accurate results.
- Agentic Retrieval: Enables LLM-guided multi-step reasoning to find complex memories.
- Multi-User Isolation: Segregates data using unique user IDs to support multi-tenant agent applications.
- Automatic Hooks: Triggers memory storage automatically upon conversation boundaries or task completion.
- Flexible Deployment: Fully compatible with Docker-based self-hosted environments for data sovereignty using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Fullrun
An AI-driven CLI tool for diagnosing, optimizing, and managing Google Ads campaigns with automated triage and guardrail protection.
Kybera: AI Wallet Control & Token Research
An AI-powered assistant for deep token research, secure wallet management, and real-time DeFi operations across multiple chains.
Snowsand Jira Cloud Integration
A comprehensive tool for AI agents to automate Jira Cloud workflows, from issue tracking to sprint management.
Image-to-Code Transformer
A high-precision utility for converting images containing text, mathematical formulas, and diagrams into structured code formats.
Weibo Hot Search Data Collector
A comprehensive multi-channel crawler and visualization engine for monitoring and archiving Weibo trending topics.
RSS Fetcher
A comprehensive RSS collection and management system designed for incremental fetching, automated tagging, and professional HTML reporting.
Featured*
