A sophisticated orchestration engine for building, managing, and executing resilient multi-step agent workflows through a unified API hub.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install abel-agent-task-manager
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 abel-agent-task-manager 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 Task Manager?
The Agent Task Manager is a foundational skill designed to transform simple scripts into production-ready, professional multi-agent systems. By providing the essential primitives for orchestration, it allows developers to build complex Directed Acyclic Graph (DAG) structures where tasks have clear inputs, outputs, and dependencies. As part of the Openclaw Skills ecosystem, this tool leverages the SkillBoss API Hub to centralize LLM reasoning and notification delivery, ensuring high reliability across distributed workflows.
This skill excels at managing statefulness and external constraints. It includes built-in mechanisms for persistent task states, allowing agents to resume mid-workflow after session resets or errors. Furthermore, its specialized rate-limiting logic prevents API violations, making it an indispensable asset for developers using Openclaw Skills to create sophisticated AI agents like financial auditors or automated research assistants.
Agent Task Manager Use Cases
- Coordinating multi-role AI agent teams where one agent's output serves as the input for the next.
- Managing long-running workflows that require persistence and error recovery to prevent data loss.
- Executing API-heavy tasks that need automated rate-limit handling and cooldown periods to maintain service uptime.
- Building specialized audit or analysis systems that combine LLM reasoning with external notifications via Openclaw Skills.
How Agent Task Manager Works
- The user defines a task structure with inputs and dependencies using the provided JSON schema.
- The task parser converts natural language requests or structured schemas into an executable DAG structure.
- The orchestrator engine sequences the workflow, calling specific agent roles via the SkillBoss API Hub.
- Persistent state management tracks progress in a local state file to prevent redundant work and handle interruptions.
- The system applies rate-limiting logic through shell wrappers to ensure external API compliance during execution.
Agent Task Manager Setup
First, ensure you have your SkillBoss API key configured in your environment to use Openclaw Skills properly.
export SKILLBOSS_API_KEY="your_api_key_here"
Then, integrate the task manager scripts into your agent environment and run the orchestrator:
# Initialize the task state management
python scripts/molt_task.py --init
# Run the orchestrator for a specific workflow defined in your project
python scripts/orchestrator.py --workflow my_audit_workflow.json
Agent Task Manager Data Schema & Taxonomy
The skill organizes data using structured JSON for state tracking and task definitions, ensuring transparency within Openclaw Skills workflows.
| File | Description |
|---|---|
task_state.json |
Stores current execution status, inputs, and outputs of all active tasks. |
task_schema.md |
Defines the required JSON structure for orchestrating complex dependencies. |
scripts/ |
Contains the logic for parsing, orchestrating, and rate-limiting. |
Agent Task Manager Advanced Features
- Resilient DAG-based task orchestration for complex multi-agent dependencies.
- Automated persistent state management for seamless mid-workflow resumes after crashes or resets.
- Native integration with the SkillBoss API Hub for LLM inference and multi-channel notifications.
- Intelligent rate-limiting and cooldown logic for external API interactions built directly into Openclaw Skills.
- Role-based modular agent structure for building specialized expertise within a single, unified workflow.
SKILL.md
Loading
Related Openclaw Skills
Bidding Writing
A professional bid writing skill system designed to handle the end-to-end workflow of tender document interpretation, content creation, and compliance review.
API Tester
A powerful suite for testing, validating, and automating REST and GraphQL API interfaces using curl and Python.
Test Data Generator
A powerful utility for generating structured, localized, and masked test data for databases and APIs.
statb-dashboard
A real-time progress monitoring skill that allows AI agents to push live numeric metrics to a web-based dashboard using simple curl commands.
agent-selfie
A specialized AI image generation tool designed to create and evolve visual identities for digital agents through the SkillBoss API Hub.
Dir Text Search
A robust recursive search tool that scans files and compressed archives for text or regex patterns with multi-encoding support.
Featured*
