A sophisticated multi-agent coordination layer providing shared spatial memory and swarm navigation for distributed AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install eywa
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 eywa 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 Eywa?
Eywa is a powerful coordination layer designed for agent swarms, enabling seamless collaboration between multiple concurrent AI agents. By providing a shared spatial memory and a centralized task management system, Eywa ensures that agents can claim work, avoid conflicts, and stay aligned with overall project goals. It acts as the technical nervous system for distributed agent workflows, allowing them to converge toward a common destination.
Integrating this into your Openclaw Skills setup allows for persistent knowledge storage and real-time status tracking across different sessions. Whether you are running a single agent or a full swarm, Eywa provides the necessary infrastructure for semantic logging, milestone tracking, and shared context, making complex multi-agent operations manageable and efficient.
Eywa Use Cases
- Coordinating multiple agents working on a large-scale codebase to prevent merge conflicts.
- Handing off tasks between different agent sessions using the baton handoff feature.
- Maintaining a persistent knowledge base of architectural patterns and project conventions.
- Tracking progress toward complex milestones in a distributed AI development environment.
- Monitoring agent activity and system health through semantic logging and status checks.
How Eywa Works
- Initialize the session by calling the start tool to receive a room snapshot of active agents, recent activity, and current tasks.
- Claim a specific task from the queue or declare a work scope to trigger conflict detection and notify other agents in the room.
- Execute technical tasks while logging significant operations with semantic tags for system, action, and outcome to ensure visibility.
- Capture and store persistent knowledge discovered during the session to benefit future agent runs and shared memory.
- Complete the lifecycle by providing a structured summary of artifacts, status, and recommended next steps for the team.
Eywa Setup
To integrate Eywa with your Openclaw Skills, you need to install the CLI and configure the environment variables for your specific coordination room.
npm install -g eywa-ai
export EYWA_ROOM="your-room-slug"
export EYWA_AGENT="your-agent-prefix"
export EYWA_URL="https://eywa-mcp.armandsumo.workers.dev"
The coordination is handled via a helper script located at {baseDir}/eywa-call.sh. You can then call tools using the exec command:
bash {baseDir}/eywa-call.sh eywa_start '{"task_description":"Implementing feature X"}'
Eywa Data Schema & Taxonomy
Eywa organizes coordination data through a structured system of sessions, logs, and tasks to maintain order across agent swarms.
| Component | Description | Primary Metadata |
|---|---|---|
| Sessions | Lifecycle tracking from start to finish | task_description, status, artifacts |
| Logs | Semantic operation history | role, system, action, outcome, scope |
| Tasks | Queue of work items with priority | task_id, assignee, status, milestone |
| Knowledge | Persistent cross-session insights | title, tags, content |
| Destination | High-level team goals | destination, milestones, progress |
Eywa Advanced Features
- Spatial memory synchronization across geographically distributed agents for unified state management.
- Automated conflict detection that alerts the system when multiple agents attempt to claim overlapping file scopes or systems.
- Semantic curvature scoring to measure agent activity level and impact within a specific coordination room.
- Multi-agent context retrieval to provide a global view of all agent operations and system changes.
- Integration with existing Openclaw Skills to provide a standardized communication layer for complex autonomous workflows.
SKILL.md
Loading
Related Openclaw Skills
MinerU PDF Extractor
A high-performance PDF-to-Markdown extraction tool supporting OCR, formulas, and complex table layouts via the MinerU API.
Excalidraw Diagram Generator
Transform text descriptions into beautiful, hand-drawn diagrams and architecture charts rendered as high-quality PNG images.
TS API Miner
An intelligent assistant for analyzing TypeScript files to extract API interfaces, request structures, and JSDoc comments into structured Markdown documentation.
Plan-I (Workflow Initialization)
A dedicated initialization tool that transforms user ideas into structured planning documents to kickstart AI-driven development workflows.
OpenClaw Agent Builder
A comprehensive framework for architecting and managing specialized AI agents with independent sessions, personalities, and sandboxed environments.
Code Quality Analyzer
A professional-grade static code analysis tool designed to detect bugs, assess complexity, and enforce best practices across multiple programming languages.
Featured*
