codex

Codex CLI Skill Guide

Available Models

Model	Best for
gpt-5.4	Flagship — complex software engineering, strongest coding + reasoning
gpt-5.4-mini	Faster/cheaper for lighter coding tasks and subagents
gpt-5.4-nano	Optimized for latency and cost on lightweight workloads
gpt-5.3-codex-spark	Near-instant real-time coding iteration (research preview)

Default recommendation: gpt-5.4 for complex tasks, gpt-5.4-mini for speed.

Running a Task

1. Ask the user (via AskUserQuestion) which model to use (default: gpt-5.4) AND which reasoning effort (xhigh, high, medium, or low) in a single prompt with two questions.
2. Select the sandbox mode required for the task; default to --sandbox read-only unless edits or network access are necessary.
3. Assemble the command with the appropriate options:
   • -m, --model <MODEL>
   • --config model_reasoning_effort="<xhigh|high|medium|low>"
   • --sandbox <read-only|workspace-write|danger-full-access>
   • --full-auto
   • -C, --cd <DIR>
   • --skip-git-repo-check
   • "your prompt here" (as final positional argument)
4. Always use --skip-git-repo-check.
5. IMPORTANT: By default, append 2>/dev/null to all codex exec commands to suppress thinking tokens (stderr). Only show stderr if the user explicitly requests to see thinking tokens or if debugging is needed.
6. Run the command, capture stdout/stderr (filtered as appropriate), and summarize the outcome for the user.
7. After Codex completes, inform the user: "You can resume this Codex session at any time by saying 'codex resume' or asking me to continue."

Quick Reference

Use case	Sandbox mode	Key flags
Read-only review or analysis	read-only	--sandbox read-only 2>/dev/null
Apply local edits	workspace-write	--sandbox workspace-write --full-auto 2>/dev/null
Permit network or broad access	danger-full-access	--sandbox danger-full-access --full-auto 2>/dev/null
Resume recent session	Inherited from original	echo "prompt" | codex exec --skip-git-repo-check resume --last 2>/dev/null
Run from another directory	Match task needs	-C <DIR> plus other flags 2>/dev/null

Following Up

• After every codex command, immediately use AskUserQuestion to confirm next steps, collect clarifications, or decide whether to resume with codex exec resume --last.
• When resuming, pipe the new prompt via stdin: echo "new prompt" | codex exec --skip-git-repo-check resume --last 2>/dev/null. The resumed session automatically uses the same model, reasoning effort, and sandbox mode from the original session.
• Restate the chosen model, reasoning effort, and sandbox mode when proposing follow-up actions.

Resuming Sessions

• Resume with new prompt: echo "follow-up prompt" | codex exec --skip-git-repo-check resume --last 2>/dev/null
• All flags must be inserted between exec and resume.
• When resuming, don't use configuration flags unless the user explicitly requests a different model or reasoning effort.

Critical Evaluation of Codex Output

Codex is powered by OpenAI models with their own knowledge cutoffs and limitations. Treat Codex as a colleague, not an authority.

Guidelines

• Trust your own knowledge when confident. If Codex claims something you know is incorrect, push back directly.
• Research disagreements using WebSearch or documentation before accepting Codex's claims. Share findings with Codex via resume if needed.
• Remember knowledge cutoffs — Codex may not know about recent releases, APIs, or changes that occurred after its training data.
• Don't defer blindly — evaluate suggestions critically, especially regarding model names, library versions, or evolving best practices.

When Codex is Wrong

1. State your disagreement clearly to the user
2. Provide evidence (your own knowledge, web search, docs)
3. Optionally resume the Codex session to discuss the disagreement. Identify yourself as Claude so Codex knows it's a peer AI discussion:

   echo "This is Claude (<your current model name>) following up. I disagree with [X] because [evidence]. What's your take?" | codex exec --skip-git-repo-check resume --last 2>/dev/null
   

4. Frame disagreements as discussions, not corrections — either AI could be wrong
5. Let the user decide how to proceed if there's genuine ambiguity

Error Handling

• Stop and report failures whenever codex --version or a codex exec command exits non-zero; request direction before retrying.
• Before using high-impact flags (--full-auto, --sandbox danger-full-access, --skip-git-repo-check) ask the user for permission using AskUserQuestion unless already given.
• When output includes warnings or partial results, summarize them and ask how to adjust using AskUserQuestion.