plan-refine-codex

Refine a Claude Code plan using OpenAI Codex. Use when you have a plan file and want a second opinion or to improve robustness.

16 stars

Best use case

plan-refine-codex is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Refine a Claude Code plan using OpenAI Codex. Use when you have a plan file and want a second opinion or to improve robustness.

Teams using plan-refine-codex should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/plan-refine-codex/SKILL.md --create-dirs "https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/main/skills/ai-agents/plan-refine-codex/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/plan-refine-codex/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How plan-refine-codex Compares

Feature / Agentplan-refine-codexStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Refine a Claude Code plan using OpenAI Codex. Use when you have a plan file and want a second opinion or to improve robustness.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

SKILL.md Source

# Plan Refinement with Codex

Use OpenAI Codex to review and refine implementation plans, adding robustness and catching edge cases.

## Prerequisites

- `codex` CLI installed and authenticated (`codex login`)
- A plan file exists (typically at `~/.claude/plans/<name>.md`)

## Usage

| Invocation | Behavior |
|------------|----------|
| `/plan-refine-codex` | Refine the current plan file (if in plan mode) |
| `/plan-refine-codex /path/to/plan.md` | Refine the specified plan file |

## How to Refine a Plan

Run codex exec with `--full-auto` and ask it to read and refine the plan file:

```bash
codex exec --full-auto "Please read the file $PLAN_FILE and refine the plan. Make it more robust, handle edge cases, and simplify where possible. Return only the improved plan in markdown format."
```

**Important**: Have codex read the file itself rather than piping content via stdin. Stdin piping has reliability issues.

## Example

```bash
codex exec --full-auto "Please read the file /Users/me/.claude/plans/my-plan.md and refine the plan. Look for edge cases like: error handling, input validation, permission issues. Return only the improved plan in markdown format."
```

## Known Issues & Workarounds

### Issue: Model not available
```
ERROR: The 'o3' model is not supported when using Codex with a ChatGPT account.
```
**Workaround**: Don't specify `-m o3`. Use the default model (`gpt-5.2-codex`).

### Issue: Stdin content not received
When piping content via stdin, codex may not receive it:
```bash
# This may fail - codex doesn't see the piped content
cat plan.md | codex exec "Refine this plan..."
```
**Workaround**: Ask codex to read the file directly in the prompt.

### Issue: Heredoc with embedded bash fails
```bash
# This fails due to nested $() and backticks
codex exec "$(cat <<'EOF'
...plan with bash code...
EOF
)"
```
**Workaround**: Reference the file path instead of embedding content.

## Iterating on Refinement

You can pass the plan to codex multiple times for progressive improvement:

1. **First pass**: Focus on edge cases and error handling
2. **Second pass**: Focus on simplification and clarity
3. **Final pass**: Focus on production-readiness

After each refinement, update your plan file with the improved version.

## Related Skills

- `/roborev-review` - Get AI code review on your implementation
- `/roborev-refine` - Automated review-fix loop

Related Skills

kitt-create-plans

16
from diegosouzapw/awesome-omni-skill

Create hierarchical project plans optimized for solo agentic development. Use when planning projects, phases, or tasks that the AI agent will execute. Produces agent-executable plans with verification criteria, not enterprise documentation. Handles briefs, roadmaps, phase plans, and context handoffs.

julien-workflow-advice-codex

16
from diegosouzapw/awesome-omni-skill

Get OpenAI Codex CLI's opinion on code, bugs, or implementation. Use when you want a second AI perspective during coding sessions.

consult-codex

16
from diegosouzapw/awesome-omni-skill

Compare OpenAI Codex GPT-5.2 and code-searcher responses for comprehensive dual-AI code analysis. Use when you need multiple AI perspectives on code questions.

codex-team

16
from diegosouzapw/awesome-omni-skill

Use when you have 2+ tasks that Codex agents should execute. Runtime-native: Codex sub-agents when available, Codex CLI fallback otherwise. Handles file conflicts via merge/wave strategies. Triggers: "codex team", "spawn codex", "codex agents", "use codex for", "codex fix".

codex

16
from diegosouzapw/awesome-omni-skill

Run OpenAI's Codex CLI agent in non-interactive mode using `codex exec`. Use when delegating coding tasks to Codex, running Codex in scripts/automation, or when needing a second agent to work on a task in parallel.

codex-sessions-skill-scan

16
from diegosouzapw/awesome-omni-skill

Daily skill health scan: analyze ~/.codex/sessions plus per-repo session logs under ~/dev (default last 1 day) and summarize skill invocations + likely failures for personal skills in ~/dev/agent-skills (missing paths, tool failures, complex-task word triggers). Optional: include best-effort local OTel signals.

codex-reviewer

16
from diegosouzapw/awesome-omni-skill

Use OpenAI's Codex CLI as an independent code reviewer to provide second opinions on code implementations, architectural decisions, code specifications, and pull requests. Trigger when users request code review, second opinion, independent review, architecture validation, or mention Codex review. Provides unbiased analysis using GPT-5-Codex model through the codex exec command for non-interactive reviews.

codex-review

16
from diegosouzapw/awesome-omni-skill

Two-pass adversarial review of design documents and implementation plans using OpenAI Codex CLI. Invokes Codex to review plans section-by-section (pass 1), then holistically (pass 2), feeding critique back for revision. Use when you have a design doc, architecture plan, or implementation plan that should be stress-tested before execution.

codex-cli-bridge

16
from diegosouzapw/awesome-omni-skill

Bridge between Claude Code and OpenAI Codex CLI - generates AGENTS.md from CLAUDE.md, provides Codex CLI execution helpers, and enables seamless interoperability between both tools

codex-advisor

16
from diegosouzapw/awesome-omni-skill

Get a second opinion from OpenAI Codex CLI for plan reviews, code reviews, architecture decisions, and hard problems. Use when you need external validation, want to compare approaches, or are stuck on a difficult problem.

business-plan

16
from diegosouzapw/awesome-omni-skill

Write, structure, and update a business plan for a solopreneur. Use when creating a plan from scratch, updating an existing plan after a pivot or new phase, or preparing a plan to share with investors, partners, or even just to clarify your own strategy. Covers executive summary, market analysis, competitive positioning, revenue model, operations plan, financial projections, and risk assessment — all adapted for a one-person business. Trigger on "write a business plan", "business plan", "create my plan", "business plan template", "update my business plan", "plan for my business", "investor pitch plan".

airtight-plans

16
from diegosouzapw/awesome-omni-skill

Write structured multi-step implementation plans in markdown format. Plans use numbered steps with clear titles and detailed instructions. Use when asked to create an implementation plan, development roadmap, or multi-step task breakdown.