A high-performance external memory plugin that connects your AI agent to a self-hosted AgentMemo server for advanced semantic retrieval.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-agentmemo
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 memory-agentmemo 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 Memory (AgentMemo)?
The Memory (AgentMemo) skill is a specialized extension designed to replace standard file-based storage with a robust semantic backend. By integrating this plugin into your Openclaw Skills ecosystem, your AI agents gain the ability to store and recall information based on meaning rather than just keyword matching. This ensures that your agents maintain deep context and continuity across long-term projects and complex workflows.
This plugin acts as a bridge to the AgentMemo server, providing a scalable way to handle high volumes of agent data. It is particularly useful for developers who want to keep their memory storage localized and secure on their own hardware while utilizing powerful vector-based search capabilities.
Memory (AgentMemo) Use Cases
- Maintaining long-term project context across multiple agent interactions.
- Building a persistent knowledge base from autonomous coding sessions.
- Implementing user-specific memory isolation in multi-tenant AI environments.
- Enhancing agent accuracy by retrieving semantically relevant historical data.
How Memory (AgentMemo) Works
- The plugin establishes a connection with a self-hosted AgentMemo HTTP server using the provided configuration.
- When autoCapture is enabled, the agent sends conversation snippets to the server for embedding and storage.
- Upon receiving a new prompt, the plugin performs a semantic search against the AgentMemo database.
- Relevant memories are fetched and injected back into the LLM context window to inform the agent's response.
- The lifecycle repeats, continuously refining the agent's available context through persistent storage.
Memory (AgentMemo) Setup
First, install the plugin using the Openclaw Skills management interface or the CLI:
openclaw install @openclaw/memory-agentmemo
Next, update your openclaw.json file with your server details:
{
"plugins": {
"memory-agentmemo": {
"serverUrl": "http://localhost:8000",
"apiKey": "your-api-key",
"userId": "dev-user"
}
}
}
Finally, ensure your AgentMemo server is running locally or on your network:
pip install agentmemo
agentmemo server --port 8000
Memory (AgentMemo) Data Schema & Taxonomy
The skill manages memory data through the following configuration schema:
| Parameter | Type | Required | Description |
|---|---|---|---|
| serverUrl | String | Yes | The URL where the AgentMemo server is accessible. |
| apiKey | String | No | Authentication token if the server is secured. |
| userId | String | No | Used to namespace memories to a specific user. |
| autoCapture | Boolean | No | If true, automatically saves new information to memory. |
| autoRecall | Boolean | No | If true, automatically retrieves memories into context. |
Memory (AgentMemo) Advanced Features
- Semantic Namespacing: Use the userId field to partition memories between different projects or users.
- Context Injection: Configurable autoRecall ensures that only the most relevant historical data occupies the context window.
- Self-Hosted Privacy: Keep all agent memory on your own infrastructure rather than third-party cloud storage.
- Extensible Logic: Works seamlessly alongside other Openclaw Skills to provide a unified AI development experience.
SKILL.md
Loading
Related Openclaw Skills
CMBC Information Assistant
A comprehensive information assistant for 民生银行 CMBC that provides real-time access to financial reports, products, and official banking resources.
Ubuntu Ollama Fleet Router
A high-performance fleet routing tool that transforms multiple Ubuntu machines into a single, load-balanced Ollama inference endpoint.
French Business Analyser
A specialized MCP skill that equips AI agents with real-time, verified French corporate and financial data from official registries.
China Stock Analysis
An intelligent stock market analysis tool designed to track, evaluate, and provide investment recommendations for Chinese and international equity markets.
Escrow MCP
A dual-deposit escrow system on Base L2 designed for secure, trustless agent-to-agent task settlement and payments.
QClaw Config Guardian
An automated configuration management skill that safeguards QClaw and OpenClaw tasks, channels, and plugin settings from data loss during upgrades.
Featured*
