A sophisticated multi-agent orchestration engine that uses a state machine to track, dispatch, and monitor complex project tasks across various AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install task-engine
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 task-engine 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 Task Engine?
Task Engine is a robust orchestration framework designed to manage multi-step technical projects by coordinating multiple AI agents. As a cornerstone of the Openclaw Skills ecosystem, it utilizes a rigorous state machine with JSON-based persistence to ensure every subtask is tracked from the initial planning phase through to final completion. By integrating agents such as Claude Code and Eva, it provides a centralized command center for complex coding and operational workflows.
This skill is essential for developers and teams looking to scale their AI-assisted development. It handles the heavy lifting of state transitions, dependency management, and progress reporting automatically. Whether you are building a new feature or managing complex system operations, Task Engine ensures that your Openclaw Skills work in perfect harmony, preventing task overlap and ensuring consistent project delivery through structured automation.
Task Engine Use Cases
- Orchestrating complex software development lifecycles across multiple specialized AI agents.
- Automated monitoring of long-running development tasks with heartbeat-based status checks.
- Managing project dependencies where specific subtasks must be verified before subsequent phases begin.
- Real-time progress tracking of AI agent activities via automated Discord notifications.
- Organizing and archiving technical project history for future reference within your Openclaw Skills environment.
How Task Engine Works
- Initialize a new project task using the create command, defining the priority level and a high-level execution plan.
- Approve the generated plan to transition the task from the PLANNING phase to the APPROVED state.
- Dispatch specific subtasks to specialized agents like Claude Code for development or Eva for system operations and testing.
- Monitor progress through automated heartbeats that update subtask status and trigger logical state transitions.
- Receive real-time alerts and digests via Discord to stay informed on task health and potential blockers.
- Transition the task through TESTING and REVIEW stages until reaching a terminal state for final completion and archival.
Task Engine Setup
To set up Task Engine, navigate to the skill directory within your workspace and ensure Python 3 is installed. For full functionality, integrate the engine into your system heartbeat to enable automated monitoring.
cd /home/zeron/.openclaw/workspace/skills/task-engine
# Create your first task to begin
python3 scripts/task_engine.py create "Feature Implementation" --priority P0 --plan "Phase 1: Architecture"
For continuous operation, add the task engine check to your heartbeat's command beat function as described in the technical documentation to fully leverage Openclaw Skills automation.
Task Engine Data Schema & Taxonomy
Task Engine organizes data using a hierarchical file structure to ensure persistence and recoverability across sessions. The system relies on the following schema:
| Component | Path / File | Description |
|---|---|---|
| Task Directories | /tasks/TASK-NNN/ |
Stores individual task.json files and assets for every project. |
| Central Index | /tasks/index.json |
A master registry used for fast lookups of all active and terminal tasks. |
| Archive Folder | /tasks/archive/ |
Permanent storage for tasks that have reached COMPLETED, FAILED, or REJECTED states. |
| Reference Docs | /references/ |
Definitions for valid state transitions and agent-specific capabilities. |
| Configuration | /config/settings.yaml |
Global settings for agent capacity limits and notification preferences. |
Task Engine Advanced Features
- Automatic state transitions that move projects to TESTING or REVIEW based on subtask completion status.
- Intelligent agent capacity management to prevent overloading specific agents like Claude Code during parallel tasks.
- Custom Discord formatting that generates visual progress bars, emoji indicators, and urgent human-ping alerts.
- Dependency tracking logic that ensures subtasks are only dispatched once their prerequisites are successfully met.
- Index reconstruction tools that can rebuild the master index from raw task files in the event of data corruption.
- Heartbeat-driven "stuck" alerts that automatically notify users if an AI agent has failed to report progress within a defined timeframe.
SKILL.md
Loading
Related Openclaw Skills
PC Control
A bridge for controlling Windows GUI environments from Linux-based AI agents using screen capture and peripheral simulation.
ComfyUI Painter
A professional-grade bridge for local ComfyUI image generation and automated CivitAI model lifecycle management within the Openclaw Skills ecosystem.
Aviation Agent
A dedicated flight briefing assistant that fetches real-time aviation weather and provides FAA regulatory guidance.
System Monitor
A specialized monitoring utility for tracking local server hardware performance including CPU, RAM, and GPU metrics.
Personalized News Collector
An intelligent agent skill that analyzes user history to deliver a prioritized, interest-based news report.
book-fetch
A specialized automation skill for searching, downloading, and archiving ebooks from Anna's Archive directly to MEGA cloud storage.
Featured*
