ccg

Combine Codex and Gemini outputs into one synthesized answer.

1|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/Leap0920/Clean-Portfolio --skill ccg-leap0920
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ccg
Source: https://github.com/Leap0920/Clean-Portfolio/tree/main/%25USERPROFILE%25/.openclaude/plugins/cache/omc/oh-my-claudecode/4.14.0/skills/ccg
Command: npx skills add https://github.com/Leap0920/Clean-Portfolio --skill ccg-leap0920

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @openai/codex, @google/gemini-cli, omc.

What problem does it solve?

It solves the problem of getting high-quality guidance by combining multiple expert perspectives into one consistent answer, reducing the chance of missing key angles or flawed assumptions.

Core Features & Use Cases

  • Tri-model orchestration (Claude + Codex + Gemini): Uses Codex for architecture/correctness/backend thinking and Gemini for UX/content clarity and alternatives, then has Claude synthesize a unified response.
  • Conflict-aware synthesis: Explicitly surfaces disagreements between Codex and Gemini and chooses a final direction with rationale and an action checklist.
  • Graceful fallback behavior: Continues with whichever external provider is available and clearly notes what perspective is missing.
  • Best use case: When you want parallel backend and UX review in one run, such as code review, cross-validation, and end-to-end product change planning.

Quick Start

Use the command /oh-my-claudecode:ccg with your task description, including that you want backend/architecture reviewed by Codex and UX/docs polish reviewed by Gemini.

Frequently Asked Questions about ccg

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

FAQPage Schema
How do I get a unified code review covering both backend architecture and UX design?

To get a unified code review, this Skill orchestrates Codex for backend architecture and Gemini for UX design, then synthesizes their outputs into a single reconciled answer. It explicitly surfaces disagreements and provides a final action checklist.

What is tri-model orchestration for software engineering tasks?

Tri-model orchestration combines Claude, Codex, and Gemini to process a single task. Claude synthesizes the outputs from Codex and Gemini, cross-validating architectural correctness and UX clarity into one cohesive response.

Do I need OpenAI Codex and Google Gemini CLI installed to use dual-advisor synthesis?

Yes, this Skill depends on @openai/codex and @google/gemini-cli. It uses omc to query these external advisors via the CLI, reading artifacts from .omc/artifacts/ask/ to build the synthesized response.

What happens if Codex or Gemini is unavailable during CLI orchestration?

If Codex or Gemini is unavailable, the Skill continues with whichever external provider is accessible. It provides a graceful fallback and clearly notes which perspective is missing from the final synthesized answer.

How do I run cross-validation for architecture and documentation improvements?

Run the command /oh-my-claudecode:ccg with your task description. Specify that you want backend architecture reviewed by Codex and documentation polish reviewed by Gemini for a synthesized cross-validation result.