ClawMemory Hub AI Skill for Openclaw

A GitHub-style version control platform for AI memories that enables agents to save, search, and fork conversation snapshots across sessions.

brettinhere

v2.3.0

Mar 12, 2026

Install

Information

Skill.md

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.


                  npx clawhub@latest install clawmemoryai

2. Manual Installation

Copy the skill folder to one of these locations

Global

~/.openclaw/skills/

Workspace

<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.


                  Help me install clawmemoryai 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 ClawMemory Hub AI Skill?

ClawMemory Hub is a comprehensive AI memory version control platform designed to provide long-term context for AI agents. By integrating this skill, developers can treat conversation history as a versioned repository, ensuring that critical knowledge is never lost and remains easily searchable. This skill allows for the creation of memory repos where snapshots are saved, committed, and shared just like source code.

The platform operates on a decentralized model where memories can be private or public, supporting collaborative environments through detailed permission management. By using this tool within the Openclaw Skills ecosystem, agents gain the ability to recall past interactions, fork existing memory sets for new projects, and maintain a high-fidelity record of complex workflows without manual documentation overhead.

ClawMemory Hub AI Skill Use Cases

Capturing long-term project history across multiple distributed AI sessions.
Collaborative memory sharing between team members using shared memory repositories.
Automatic snapshotting of complex coding or research workflows to prevent context drift.
Securely storing and searching past agent interactions while maintaining original conversation fidelity.
Forking successful memory templates to jumpstart new agent personalities or project contexts.

How ClawMemory Hub AI Skill Works

The skill initializes by loading the default memory repository summary and injecting it into the system prompt at the start of every session.
Conversation turns are buffered locally in the session memory to maintain low latency.
An automatic trigger silently commits snapshots to the cloud every 10 turns to ensure data persistence.
Users can manually trigger a save at any time using natural language commands like save memory.
The skill interfaces with the ClawMemory Hub API to manage repositories, handle forks, and sync collaborator permissions.

ClawMemory Hub AI Skill Setup

First, generate your API key at the ClawMemory Hub settings page. Then, run the setup script and configure your credentials.

# Run the initial setup script
bash ~/.openclaw/workspace/skills/clawmemoryai/scripts/setup.sh

Edit your configuration file at ~/.clawmemoryai/config.json with the following structure:

{
  "baseUrl": "https://clawmemory.ai",
  "apiKey": "cmh_your_api_key_here",
  "defaultUser": "your_username"
}

ClawMemory Hub AI Skill Data Schema & Taxonomy

ClawMemory Hub organizes data into structured commits within repositories. Each commit preserves the exact conversation turns and relevant metadata.

Component	Description
Repositories	Named containers for memory (e.g., username/project-name)
Commits	Snapshots containing an array of message turns with timestamps
Metadata	Includes model name (e.g., Claude 3.5), duration, and agent ID
Privacy	Repositories are private by default but can be toggled to public

ClawMemory Hub AI Skill Advanced Features

Silent background auto-saving every 10 turns to ensure no context is lost during long sessions.
Intelligent server-side redaction that automatically masks sensitive credentials like OpenAI keys or GitHub tokens.
Granular collaborator management supporting read, use, and write permission levels.
Contextual switching allows agents to pivot between different project memories instantly via CLI or natural language.
Support for multi-agent synchronization through shared cloud-based memory repos within the Openclaw Skills framework.