codex

Wraps the OpenAI Codex CLI to run independent code reviews, adversarial challenges, and consult sessions.

Updated Jun 16, 2026
One-click install
npx skills add https://github.com/adamtpang/summon.company --skill codex-adamtpang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/adamtpang/summon.company/tree/main/.claude/skills/codex
Command: npx skills add https://github.com/adamtpang/summon.company --skill codex-adamtpang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Getting a second opinion on code usually means switching tools and losing context. This Skill brings the OpenAI Codex CLI into your current session so you can request an independent diff review, an adversarial challenge of your implementation, or an open-ended consultation without leaving your workflow. ## Core Features & Use Cases - Independent Code Review: Runs codex review on your diff with a pass/fail gate, giving you an outside perspective before shipping. - Adversarial Challenge Mode: Instructs Codex to actively try to break your code, surfacing edge cases and design weaknesses you may have missed. - Consult with Session Continuity: Ask Codex anything and keep follow-up questions in the same session, useful for architecture debates or debugging discussions. - Use Case: Before merging a feature branch, invoke the skill to run a Codex review of the diff, then use challenge mode to probe the riskiest logic, and only proceed once both pass. ## Quick Start Ask the assistant to run a codex review of the current branch diff and report any issues it finds.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I get a second opinion on a code diff with Codex?

Invoke the skill and choose review mode, which runs codex review on your current diff with a pass/fail gate. The result is an independent assessment you can compare against your own review before merging.

What is the difference between codex review and codex challenge modes?

Review mode performs a standard independent diff review with a pass/fail outcome. Challenge mode is adversarial: it instructs Codex to actively try to break your code, exposing edge cases and design flaws rather than just checking correctness.

Does the codex skill support follow-up questions in the same session?

Yes, consult mode maintains session continuity so you can ask follow-up questions without restating context. This is useful for extended architecture or debugging conversations with Codex.

Can the codex skill run in headless or CI environments?

Yes, the preamble detects headless, spawned, and interactive session kinds. In headless mode it blocks instead of prompting, and in spawned sessions it auto-chooses recommended options and reports results as prose.

What happens if AskUserQuestion is unavailable when the codex skill runs?

The skill follows a documented fallback: headless sessions stop with a BLOCKED status, interactive sessions render the decision brief as prose and wait for a typed reply, and spawned sessions auto-select the recommended option.