Shared Memory for Openclaw

A unified, persistent memory system designed for cross-team, cross-window, and cross-session AI agent collaboration.

davidme6
v1.0.0
Mar 30, 2026
0
696
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install vc-shared-memory

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 vc-shared-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 Shared Memory?

Shared Memory is a robust state management and communication framework tailored for virtual companies and autonomous AI teams. It serves as a centralized nervous system that breaks down information silos, allowing agents to access a shared pool of corporate announcements, project workspaces, and historical knowledge. By integrating this into Openclaw Skills, developers can ensure their agents maintain a consistent context regardless of which session or window they are operating in.

The skill provides a structured way to handle organizational data, ranging from high-level CEO decisions to granular employee status updates. This persistent layer is essential for multi-agent workflows where context continuity and real-time coordination are critical for project success.

Shared Memory Use Cases

  • Establishing a centralized company board for broadcasting strategic decisions across all AI agents.
  • Managing complex multi-team projects by tracking milestones and tasks in a shared workspace.
  • Building an institutional knowledge base to store lessons learned and technical best practices.
  • Tracking agent workload and availability to optimize task allocation in a virtual company environment.
  • Synchronizing information across disparate agent sessions and windows to ensure data consistency.

How Shared Memory Works

  1. Initialize the memory pool structure in the local directory to create a persistent storage environment.
  2. Agents utilize the Write mechanism to contribute announcements, project updates, or status changes to JSON-based storage.
  3. The system automatically generates unique IDs, timestamps, and metadata for every entry to ensure data integrity.
  4. Agents perform targeted Reads or full-text Searches to retrieve relevant context based on priority or category.
  5. Automated notification triggers alert specific agents or teams when critical updates, such as project blockers, are recorded.

Shared Memory Setup

To get started with this skill within the Openclaw Skills ecosystem, follow these installation steps:

# Navigate to your skill directory
cd shared-memory

# Install required dependencies
npm install

# Initialize the memory pool and check statistics
node cli.js stats

# View current company announcements
node cli.js announce list

Shared Memory Data Schema & Taxonomy

The skill organizes data using a hierarchical directory structure under ~/.shared-memory/. Key data models include:

Data Type Description Key Fields
Announcements Strategic notices and alerts id, type, priority, visibleTo
Projects Active collaboration spaces teams, members, timeline, tasks
Knowledge Persistent institutional memory category, learnedBy, importance
Status Real-time agent availability currentTask, workload, availability

Shared Memory Advanced Features

  • Role-based access control (RBAC) to define visibility and edit permissions for specific agents or teams.
  • Automated audit logging for every interaction, providing a complete history of memory modifications.
  • Seamless integration with virtual-company workflows for automatic task recording and lesson harvesting.
  • Built-in CLI tools for administrative tasks, including manual data entry, search, and system backups.
  • Smart notification hooks that can be extended to support external messaging platforms or inter-agent signals.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*