Harness Subagent

harness-subagent

**Dispatch another coding-agent CLI as a one-shot subagent, then synthesize.**

[](LICENSE) [](https://agentskills.io) [](https://code.claude.com/docs/en/headless) [](https://github.com/openai/codex) [](https://docs.x.ai)

Ever wanted to call **Codex** from **Claude Code**? Ever wanted **Grok Build** to invoke **Claude Code**? Ever wanted **Claude Code** to send a diff to **GPT** for a second opinion?

That's this skill. Stay in the parent you are already in — Cursor Agent, **cursor-agent**, Claude Code, Codex, Grok Build, or **Grok Bot**. It writes a brief, dispatches another coding-agent harness as a **one-shot subagent**, then the parent synthesizes.

The other harness is not an oracle. A model reviewing its own work reproduces its own blind spots; a differently-trained harness does not. That worth is destroyed the moment you forward its answer without judging it.

This is a [Claude skill](https://code.claude.com/docs/en/skills) in the [Agent Skills](https://agentskills.io) format. The same protocol works from any agent that can run a CLI in the background.

Why I use it this way

These are **my** defaults, not the protocol. The skill will not pick a harness from the job type. You name the CLI, or you pin defaults in a **user config file** the parent is instructed to read, or it asks once.

I find **Opus** (Claude Code) superior at **UI work** — layout, interaction, the thing on the screen. I find **GPT / Codex** superior at **reviewing**, especially **visual review**: screenshots plus the named CSS/JS, then a verdict on whether the defect is real.

So my default loop is: ship the UI with Opus, capture