A robust collaboration framework for OpenClaw agents to delegate tasks and maintain state across multi-agent workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cross-agent-notify
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 cross-agent-notify 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 Cross-Agent Notification Protocol?
The Cross-Agent Notify protocol provides a standardized workflow for multi-agent environments using Openclaw Skills. It addresses the critical challenge of session timeouts in complex agentic chains by utilizing a persistent session layer (agent:
By establishing a four-step mandatory sequence—task file updates, inbox messaging, wakeup calls, and group tracing—this skill enables developers to build reliable, transparent, and traceable AI agent networks. It acts as the communication backbone for any multi-agent architecture where agents must hand off work or request assistance from specialized peers.
Cross-Agent Notification Protocol Use Cases
- Delegating specific technical tasks to specialized sub-agents.
- Handing off a project from a researcher agent to a writer agent after data collection.
- Waking up an idle agent to provide real-time input or peer review.
- Maintaining a centralized audit trail for agent-to-agent interactions in enterprise environments.
How Cross-Agent Notification Protocol Works
- Task Metadata Update: The requesting agent modifies the shared task file, updating the owner and the communication fields to define the reply path.
- Inbox Notification: A structured text file is written to the target agent's dedicated inbox directory to provide full context and instructions.
- Targeted Wakeup: A sessions_send command is issued to the target agent's persistent session layer, ensuring the message is received regardless of the agent's current state.
- Trace Logging: A concise trace message is posted to the shared Feishu group to provide human visibility and a searchable history of the collaboration event.
Cross-Agent Notification Protocol Setup
To implement this protocol within your Openclaw Skills environment, ensure your agents are configured with access to the shared inbox directory. Use the following code structure for sending wakeups:
sessions_send(
sessionKey=None,
agentId="target-agent-id",
message="task TASK-123 assigned to you, check inbox"
)
Ensure the directory structure shared/inbox/<target_agent>/ is initialized before use.
Cross-Agent Notification Protocol Data Schema & Taxonomy
This protocol utilizes standardized metadata fields within task files and a specific directory taxonomy for Openclaw Skills communication.
| Field | Description | Requirement |
|---|---|---|
owner |
The ID of the agent currently responsible for the task. | Mandatory |
communication.reply_to_agent |
The ID of the agent that should receive results upon completion. | Mandatory |
communication.submitter_agent |
The ID of the agent that initiated the request. | Mandatory |
Inbox File Convention: shared/inbox/<target_agent>/YYYY-MM-DD-HHMM-task-description.txt
Cross-Agent Notification Protocol Advanced Features
- Persistent session activation via agent:
:main to bypass session timeout limitations. - Standardized collaboration chains that integrate file-based messaging with real-time session wakeups.
- Dynamic routing rules that prevent hardcoding agent names and support flexible task ownership transfers.
- Automated Feishu group integration for real-time monitoring of agent-to-agent handoffs.
SKILL.md
Loading
Related Openclaw Skills
Guoshun Digital Team Collaboration Skill
A professional Multi-Agent collaboration framework for OpenClaw that simulates a digital team's division of labor and structured discussion mechanisms.
Adyen Data Protection API Skill
A specialized automation interface for managing Adyen Data Protection requests and programmatically triggering Subject Erasure.
Adyen Checkout API
A comprehensive AI agent skill for managing Adyen Checkout API workflows, including payments, sessions, and transaction management.
Adyen Payout API Skill
A specialized AI agent skill designed to automate and manage the Adyen Payout API lifecycle, covering everything from instant payouts to third-party confirmations.
Adyen Payment API Skill
A specialized Openclaw Skills integration for the Adyen Payment API v68, enabling AI agents to automate transaction lifecycles and secure payment processing.
Adyen BinLookup API Skill
A high-performance interface for Adyen BinLookup services, enabling AI agents to verify 3D Secure status and calculate transaction cost estimates.
Featured*
