A sophisticated coordination framework for managing multi-agent development pipelines through shared JSON task files and centralized AGI dispatch.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install team-tasks-skip
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 team-tasks-skip 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 Team Tasks?
Team Tasks is a powerful orchestration skill designed to synchronize specialized AI agents within the Openclaw Skills ecosystem. It serves as a centralized command center, enabling an AGI to dispatch specific workloads to code-agents, test-agents, and docs-agents without requiring direct agent-to-agent communication. This ensures a structured, manageable, and traceable development lifecycle for any project size.
By utilizing shared JSON task files, this skill maintains a persistent state of project progress, allowing for both simple sequential workflows and complex parallel execution. Team Tasks provides the necessary infrastructure to scale AI-driven development by providing clear status tracking, result collection, and automated dispatch logic that integrates seamlessly with existing communication channels.
Team Tasks Use Cases
- Managing sequential development pipelines such as Code to Test to Documentation.
- Orchestrating parallel workstreams where multiple agents work on independent features simultaneously.
- Coordinating complex dependencies using Directed Acyclic Graphs (DAG) for large-scale software projects.
- Tracking the status and output of multi-agent tasks in a centralized, persistent log.
- Automating the hand-off of context and results between specialized agents to ensure project continuity.
How Team Tasks Works
- Initialize a project using the task manager CLI, selecting either linear or DAG mode to define the workflow structure.
- Assign specific task descriptions and agent roles to the project stages to define the work scope.
- Identify the next available or ready tasks based on current progress or dependency fulfillment.
- Dispatch tasks to the assigned agents using session keys, providing the necessary project context and previous results.
- Capture and save the agent's output using the result command to maintain the inter-agent context chain.
- Update the task status to done, which automatically unblocks the next stage or parallel tasks in the pipeline.
Team Tasks Setup
To begin coordinating your agents with this Openclaw Skills component, use the task manager script located in your skill directory:
# Initialize a sequential project
python3 scripts/task_manager.py init my-project -g "Build a REST API" -p "code-agent,test-agent"
# Assign a task to the first agent
python3 scripts/task_manager.py assign my-project code-agent "Implement the core API endpoints"
# Check project status
python3 scripts/task_manager.py status my-project
Team Tasks Data Schema & Taxonomy
The skill manages data within the /home/ubuntu/clawd/data/team-tasks/ directory, using JSON files for project state. The schema includes:
| Attribute | Type | Description |
|---|---|---|
project |
String | The unique project identifier (slug). |
goal |
String | The primary objective of the multi-agent effort. |
mode |
String | The execution logic (linear or dag). |
status |
Enum | Project-wide status: pending, in-progress, done, failed. |
tasks |
Object/Array | Detailed task definitions including agent assignments, dependencies, and results. |
logs |
Array | A chronological history of project events and debug information. |
Team Tasks Advanced Features
- Mode B (DAG) Support: Define complex task dependencies to enable sophisticated parallel processing.
- Parallel Dispatch: Automatically identify and launch all tasks whose dependencies have been satisfied.
- Dependency Forwarding: Pass outputs from multiple completed predecessors directly into the context of a new task.
- Cycle Detection: Integrated validation to prevent circular dependencies in complex DAG configurations.
- Targeted Retries: Reset specific failed stages without affecting the progress of unrelated parallel branches.
- Visualization Tools: Generate tree views of task dependencies and current status icons via the CLI.
SKILL.md
Loading
Related Openclaw Skills
n8n Workflow Management
A comprehensive management tool for n8n to programmatically create, test, and optimize automated workflows.
Loyalty Designer
A strategic framework for designing margin-positive loyalty programs that increase customer lifetime value through points, tiers, and referral mechanics.
Amarin Memory
A lightweight, SQLite-powered memory system for AI agents that handles long-term storage, semantic retrieval, and natural memory decay.
Apple Silicon AI Fleet Manager
A high-performance local AI inference engine that aggregates the compute power of M-series Macs into a unified, scalable cluster.
Lawyer Six Steps
A professional legal assistant skill implementing the Chinese Lawyer Six-Step Method to generate comprehensive case analysis reports.
Gemini CLI Integration
A bridge for utilizing the local Gemini CLI to perform secondary model reasoning, content generation, and structured data tasks on Linux hosts.
Featured*
