configure-subagents

Configure model and reasoning effort parameters for Oh No Harness Claude Code subagents.

10|Updated May 9, 2026

Search Tags:#automation #workflow #subagents #configuration #claude-code #agent-tuning

Ready to run

npx skills add https://github.com/jcwleo/oh-no-harness --skill configure-subagents

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: configure-subagents
Source: https://github.com/jcwleo/oh-no-harness/tree/main/plugins/oh-no-harness/skills-claude/configure-subagents
Command: npx skills add https://github.com/jcwleo/oh-no-harness --skill configure-subagents

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill solves the challenge of inconsistent AI agent performance by allowing users to explicitly configure the model and reasoning effort for each of the nine Oh No Harness subagents.

Core Features & Use Cases

Granular Control: Set specific models and reasoning effort levels for individual roles like planner, executor, and verifier.
Environment Optimization: Manage diversity settings and CLIProxyAPI integration to ensure high-quality, reliable agent outputs.
Use Case: A developer needs to ensure that the planner and code-reviewer agents use high-effort models to handle complex architectural changes, while other agents use faster, more efficient models to maintain workflow speed.

Quick Start

Run the configure-subagents skill to launch the interactive setup wizard for your Claude Code subagents.

Frequently Asked Questions about configure-subagents

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

FAQPage Schema

How do I configure Claude Code subagent models and reasoning effort?▼

You can configure Claude Code subagent models and reasoning effort by running the interactive setup wizard, which lets you explicitly set parameters for each of the nine Oh No Harness subagents.

Can I set different reasoning effort levels for individual Claude Code agents?▼

Yes, you can set specific models and reasoning effort levels for individual Claude Code agents, allowing roles like planner and verifier to use high-effort models while others use faster, more efficient models.

Do I need Oh No Harness to manage Claude Code subagent settings?▼

Yes, you need an active Claude Code environment and the Oh No Harness plugin installation to apply configuration changes atomically and manage environment-specific agent behavior.

What is the best way to optimize Claude Code agent diversity and proxy integration?▼

The best way to optimize Claude Code agent diversity and proxy integration is to use the configuration wizard to manage CLIProxyAPI settings and diversity parameters, ensuring high-quality, reliable agent outputs.

Why does my Claude Code subagent configuration fail to apply atomically?▼

Atomic configuration application fails when the active Claude Code environment is missing or the Oh No Harness plugin is not properly installed, preventing environment-specific agent behavior changes.