subagent-to-subharness

Translate host tasks into subharness dispatch decisions with role mapping.

4|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/danmestas/wardrobe --skill subagent-to-subharness
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-to-subharness
Source: https://github.com/danmestas/wardrobe/tree/main/skills/subagent-to-subharness
Command: npx skills add https://github.com/danmestas/wardrobe --skill subagent-to-subharness

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams replace ad-hoc subagent dispatch with deterministic subharness orchestration by translating host task shapes into the appropriate harness role and framing the work in caveman-standard prompts.

Core Features & Use Cases

  • Translate host tasks into subharness dispatch decisions mapped to backend roles.
  • Frame tasks in caveman-standard prompts to guide subharness execution.
  • Read worker output and route next steps based on results.
  • Provide deterministic task framing and role mapping with clear fail-safes for common edge cases.
  • Clarify when to spawn a subagent versus a subharness depending on context and cost considerations.

Quick Start

Provide a host task description and this skill will map it to the correct subharness workflow.

Frequently Asked Questions about subagent-to-subharness

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

FAQPage Schema
How do I route tasks to subagents instead of using ad-hoc dispatch?

Subharness orchestration replaces ad-hoc dispatch by translating host task shapes into deterministic backend role mappings, using caveman-standard prompt framing to guide worker execution and route next steps based on output.

What is the difference between spawning a subagent and using a subharness?

Spawning a subagent is an ad-hoc dispatch method, while a subharness applies deterministic task framing, role mapping, and output-driven workflow routing with clear fail-safes for edge cases.

How do I map a host task to the correct subharness workflow?

Provide a host task description to translate the task shape into the appropriate subharness dispatch decision, mapping it to the correct backend role and framing the work in caveman-standard prompts.

Does subharness orchestration work with claude-code backends?

Yes, subharness orchestration explicitly supports claude-code and related backends, translating host tasks into backend role mappings and routing next steps based on worker output.

When should I not use deterministic task routing for subagents?

Avoid deterministic task routing when ad-hoc subagent dispatch is sufficient, or when cost considerations favor spawning a subagent over the structured role mapping and fail-safes of a subharness workflow.