A comprehensive process framework for building and managing self-sustaining AI agent teams with automated work loops.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-team-kit
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 agent-team-kit 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 Agent Team Kit?
The Agent Team Kit provides a robust structural framework for organizing multiple AI agents into a cohesive, self-operating unit. By leveraging specialized Openclaw Skills, this kit eliminates human bottlenecks through automated intake loops and clearly defined role ownership. It transforms static agent interactions into a dynamic ecosystem where agents proactively discover, triage, and execute tasks based on a shared heartbeat.
This skill is essential for developers looking to scale their AI operations without increasing manual oversight. Using Openclaw Skills like this ensures that your agent teams remain productive, organized, and capable of handling complex workflows through specialized execution roles and a standardized communication protocol. It shifts the burden of management from the human user to a self-correcting system of roles and responsibilities.
Agent Team Kit Use Cases
- Managing complex multi-agent projects where tasks need to be distributed dynamically without constant human direction.
- Automating the discovery and triage of new project opportunities to ensure a continuous pipeline of work.
- Scaling AI operations by spawning specialized execution agents for specific tasks like design, architecture, or coding.
- Maintaining project momentum in Openclaw Skills environments through automated heartbeat checks and status tracking.
- Implementing a self-service work queue where agents autonomously pick up ready tasks based on their roles.
How Agent Team Kit Works
- Discover: The Scout role identifies new opportunities or needs and logs them into the central opportunities file.
- Triage: The Rhythm role evaluates raw ideas, prioritizes them, and moves them into the backlog.
- Ready: Tasks move into a self-service queue where they are marked as ready for any available agent to claim.
- Execute: Assigned execution agents perform the technical work using their specific Openclaw Skills and specialized toolsets.
- Feedback: Upon completion, the agent logs results which triggers a feedback loop to generate new ideas or refine the process.
Agent Team Kit Setup
To begin, copy the process templates into your workspace root to establish the team structure:
cp -r skills/agent-team-kit/templates/process ./process
Next, integrate the heartbeat configuration to enable proactive agent operations. If you have an existing heartbeat file, append the template content:
cat skills/agent-team-kit/templates/HEARTBEAT.md >> HEARTBEAT.md
Finally, customize the process/ROLES.md file to match your specific team requirements and define the human lead's strategic oversight role. This ensures your Openclaw Skills are properly aligned with your project goals.
Agent Team Kit Data Schema & Taxonomy
The kit organizes team data and state across several specialized Markdown files within the process directory:
| File | Purpose | Primary Owner |
|---|---|---|
INTAKE.md |
Reference guide for the 5-phase operational loop | System |
ROLES.md |
Definitions of missions, ownership, and cadences | Human Lead |
OPPORTUNITIES.md |
Storage for raw discoveries and unrefined ideas | Scout |
BACKLOG.md |
Prioritized queue of tasks ready for execution | Rhythm |
STATUS.md |
Real-time tracking of which agent is doing what | All Agents |
HEARTBEAT.md |
Trigger list for proactive health and status checks | System |
This schema ensures that all activity within your Openclaw Skills ecosystem is documented and actionable.
Agent Team Kit Advanced Features
- Heartbeat-driven proactive triggers that automatically nudge stalled tasks or alert roles when queues are empty.
- Dynamic specialist spawning allows the main agent to coordinate while sub-agents handle specific execution tasks.
- Metrics tracking for cycle time and queue depth to monitor the overall health of your Openclaw Skills workflow.
- Customizable intake loops that allow for the addition of validation gates or parallel execution tracks.
- Anti-pattern detection mechanisms designed to prevent human bottlenecks and ensure agents don't wait for permission to work.
SKILL.md
Loading
Related Openclaw Skills
Agent Memory Kit
A modular framework that prevents AI context loss by organizing memory into episodic, semantic, and procedural layers.
Agent Identity Kit
A portable identity system for creating, validating, and publishing standardized JSON identity cards for AI agents.
Agent Bridge Kit
A cross-platform bridge for AI agents to interact with social networks, news feeds, and skill directories through a single configuration.
Agent Autonomy Kit
A strategic framework designed to transform AI agents from reactive responders into proactive task-executors that operate independently.
mcd-cn McDonald's China Assistant
A CLI-based MCP client for McDonald's China that enables campaign calendar tracking, coupon lookups, and automated coupon claiming.
Yijing Divination AI Skill
An advanced I Ching divination engine that combines traditional coin and yarrow methods with poetic, AI-driven Oracle interpretations for deep introspection.
Featured*
