decompose-mission

Decompose multi-step instructions into executable tasks for robot agents.

Updated Jul 5, 2026
One-click install
npx skills add https://github.com/bensonlee5/openral --skill decompose-mission
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: decompose-mission
Source: https://github.com/bensonlee5/openral/tree/main/rskills/decompose-mission
Command: npx skills add https://github.com/bensonlee5/openral --skill decompose-mission

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the decomposition of multi-step instructions into a sequence of executable tasks, enhancing task planning and execution in robot agents.

Core Features & Use Cases

  • Task Decomposition: Breaks down complex instructions into manageable subtasks with verifiable done-conditions.
  • Sequential Execution: Executes and verifies subtasks in an ordered sequence, with replanning on failure.
  • Embodiment-Agnostic: Works with any robot embodiment through a flexible rSkill packaging format.

Quick Start

Use the decompose-mission skill to plan and execute a multi-step instruction: decompose-mission --instruction "stack the bowls in the drawer and then place the plate on the cookie box".

Frequently Asked Questions about decompose-mission

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate task decomposition for multi-step robot instructions?

Automating task decomposition for multi-step robot instructions requires breaking down commands into a sequence of executable subtasks. This skill handles it by parsing instructions into ordered steps with verifiable done-conditions for robot agents.

What is sequential execution planning in robotics and how does replanning work?

Sequential execution planning in robotics executes subtasks in order and verifies completion. If a subtask fails, replanning is triggered to adjust the remaining sequence, ensuring the overall instruction still succeeds.

Do I need specific tools to use this task planning Skill?

Yes, this task planning Skill requires an S2 decision procedure and specific tools: execute_rskill, query_scene, query_task_progress, memory_write, memory_search, and emit_prompt to automate execution and failure handling.

Can I use execution automation with any robot embodiment?

Yes, execution automation works with any robot embodiment. It uses a flexible rSkill packaging format to ensure task decomposition and execution planning remain hardware-agnostic across different robotic agents.

What happens when a subtask fails during sequence execution?

When a subtask fails during sequence execution, the Skill triggers replanning to adjust the subsequent tasks. This failure handling mechanism ensures the robot agent can recover and continue the mission.

What's the best way to plan a complex instruction like stacking bowls and plates?

The best way to plan complex instruction stacking is to pass the full command to the Skill. It decomposes the multi-step instruction into ordered subtasks, each with verifiable done-conditions for automated execution.