Routing Work Across AI Harnesses

What this Claude Code skill does

Designs hybrid agent workflows that route tasks intelligently between AI coding agent harnesses based on task characteristics rather than defaulting to one tool for everything. Use when a team uses multiple AI coding agents, when deciding which harness handles which task type, or when designing handoff protocols between agents with different architectural dispositions. Common phrases: "which AI co

Routing Work Across AI Harnesses is a packaged skill for Claude Code: a SKILL.md plus references and a runnable test that drops into .claude/skills/, so your agent can run this workflow reliably instead of you re-explaining it each time. Install it via MCP, run it as a CLI, or fork it into your own.

Inside the SKILL.md

<h1>Routing Work Across AI Harnesses</h1>
<h2>Why Routing Matters</h2>
<p>The era of picking one AI coding tool is ending. Developers extracting the most
value today use multiple harnesses and route work based on what each task needs.</p>
<p>The skill is not in mastering either tool. It is knowing which harness's architectural
disposition matches the kind of work you are doing right now.</p>
<blockquote>
<p>"He uses Claude Code for planning, orchestrating his terminal, and explaining how
parts of the codebase work... Codex is for the actual code because the Codex code
just straight up has fewer bugs."</p>
</blockquote>
<p>This is not a preference — it reflects a structural difference in what each harness
is architecturally suited for.</p>
<hr />
<h2>The Two Harness Dispositions</h2>
<p>Before routing, understand the dispositions:</p>
<table>
<thead>
<tr>
<th>Disposition</th>
<th>Characteristics</th>
<th>Suited For</th>
…