Stats
Actions
Tags
From openai-codex
Consult OpenAI Codex for a second opinion when stuck. Use when you've tried 2+ approaches that failed, need input on architecture decisions, or want security review.
How this skill is triggered — by the user, by Claude, or both
Slash command
/openai-codex:codex-cliThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Get a second opinion from Codex (OpenAI's GPT CLI) when stuck.
Get a second opinion from Codex (OpenAI's GPT CLI) when stuck.
DO consult when:
DON'T consult when:
Good prompt includes context:
Context: [What user is working on]
What's been tried: [Approaches attempted, why they failed]
Question: [Specific question, not vague "help me"]
Relevant code: [Key snippets if applicable]
Always wait for codex to respond. It can take anywhere from 1-20 minutes. Always wait for the full response
DO:
DON'T:
npx claudepluginhub stevennevins/openai-codex --plugin openai-codexGets a second opinion from OpenAI Codex CLI during planning to validate approaches, identify risks, and suggest alternatives.
Proactively suggests second opinions from other LLMs on architectural decisions, design trade-offs, critical code reviews, and security-sensitive logic.
Consults Gemini 2.5 Pro, OpenAI Codex, and Claude for second opinions on debugging failures, architectural decisions, security validation, and fresh perspectives.