A local memory service that allows AI agents to persist learnings, preferences, and goals across sessions using secure Ed25519 cryptography.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install persistent-private-agent-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-private-agent-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 Agent Memory Service (Improved)?
The Agent Memory Service (Improved) is a robust solution for developers looking to add long-term memory to their local AI agents. By leveraging the Openclaw Skills framework, this tool ensures that your agents can recall user preferences, track learning histories, and maintain goal progress across restarts. Unlike basic storage solutions, this version features advanced Ed25519 cryptographic signatures and BIP39 key derivation, ensuring that your agent identity and data remain secure and authenticated within your local environment.
This skill addresses the common problem of context loss in AI interactions. By providing a dedicated service to store and retrieve structured snapshots, Openclaw Skills allow for a more personalized and intelligent agent experience that evolves over time without relying on expensive or privacy-invasive cloud providers.
Agent Memory Service (Improved) Use Cases
- Maintaining user-specific communication styles and technical preferences across multiple agent restarts.
- Tracking long-term project goals and asynchronous task progress for complex development workflows.
- Building a persistent repository of lessons learned to prevent the agent from repeating previous errors.
- Identifying and tracking technical knowledge gaps to streamline learning and resource gathering.
- Securely storing agent state in local environments without relying on third-party cloud storage.
How Agent Memory Service (Improved) Works
- Initialize the local service and database environment using the provided setup scripts.
- Generate a unique agent identity derived from a BIP39 recovery phrase to ensure cryptographic security.
- Register the agent identity with the local service using Ed25519 signatures for verified authentication.
- Capture structured snapshots of agent state, including learning events and active goals during active sessions.
- Sign and store memory data via the CLI client, which ensures only authorized agents can modify their memory records.
- Retrieve and reload previous memory snapshots at the start of new sessions to restore full context instantly.
Agent Memory Service (Improved) Setup
Install the necessary Python dependencies to ensure the service runs correctly:
pip install fastapi uvicorn psutil pydantic cryptography mnemonic base58
Run the automated setup and start the service to integrate with your Openclaw Skills:
# Run the setup script
./scripts/setup.sh
# Start the local memory service
./scripts/start.sh
Register your agent identity to begin using the memory management features:
./scripts/memory_client.py register
Agent Memory Service (Improved) Data Schema & Taxonomy
The memory service organizes data using a structured JSON schema to ensure compatibility across different Openclaw Skills implementations.
| Component | Description | Key Attributes |
|---|---|---|
| user_preferences | Tracks interaction styles | category, key, value, confidence |
| learning_history | Logs successes and failures | event_type, lesson_learned |
| knowledge_gaps | Identifies missing information | topic, priority |
| active_goals | Monitors long-term tasks | title, progress |
All data is stored locally in an SQLite database (agent_memory.db) to maintain privacy and performance.
Agent Memory Service (Improved) Advanced Features
- Ed25519 Cryptographic Security: Features a high-performance signature implementation for tamper-proof memory storage and retrieval.
- BIP39 Identity Management: Uses standard recovery phrases to derive agent keys, ensuring identity portability and secure backups.
- Local-Only Infrastructure: Zero cloud dependencies, ensuring complete data privacy and low-latency retrieval for all your Openclaw Skills.
- Structured JSON Snapshots: Facilitates easy integration with other agents for complex multi-agent workflows and shared learning states.
- Automated Authentication: The client handles all signing operations automatically, making secure storage seamless for the developer.
SKILL.md
Loading
Related Openclaw Skills
Windsurf Cascade Agent Skill
A comprehensive AI agent skill for Windsurf IDE that enables autonomous coding, multi-file editing, and complex terminal-based workflows.
Skill Creator
A comprehensive framework for designing, initializing, and packaging modular skills that provide AI agents with specialized procedural knowledge.
Botbook.space Relationships
A comprehensive social graph skill for AI agents to establish connections, manage relationship types, and engage with a bot-centric community.
Botbook.space Social Network for AI Agents
A complete REST API-driven social platform where AI agents post updates, follow peers, and build autonomous communities.
Docker Sandbox
A powerful tool for creating and managing isolated Docker VM sandboxes to safely execute AI agents and untrusted code.
Cloudflare DNS and Tunnel Management
A comprehensive automation skill for managing Cloudflare DNS records, domain zones, and secure tunnels via the Cloudflare API.
Featured*
