ccg

Orchestrate parallel Codex and Gemini advisor prompts and synthesize their outputs.

1|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/atmigtnca/sidep-ops --skill ccg-atmigtnca
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ccg
Source: https://github.com/atmigtnca/sidep-ops/tree/main/skills/ccg
Command: npx skills add https://github.com/atmigtnca/sidep-ops --skill ccg-atmigtnca

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates parallel advisor prompts from Codex and Gemini and synthesizes their outputs to provide a consolidated answer, reducing manual coordination and conflicting results.

Core Features & Use Cases

  • Parallel advisor prompts: runs Codex and Gemini prompts in tandem and collects their outputs.
  • Synthesis & validation: Claude combines the two perspectives into a single, actionable recommendation.
  • Artifact generation: artifacts are stored under .omc/artifacts/ask/ for traceability and audit.

Quick Start

Invoke /ccg with a task description to run Codex and Gemini prompts in parallel and synthesize a final recommendation.

Frequently Asked Questions about ccg

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

FAQPage Schema
How do I synthesize parallel outputs from Codex and Gemini for architecture tasks?

Synthesizing parallel outputs from Codex and Gemini is done by invoking the /ccg command, which orchestrates parallel advisor prompts and combines their perspectives into a single actionable recommendation.

What is cross-validation between Codex and Gemini in AI model orchestration?

Cross-validation between Codex and Gemini is a process where parallel advisor prompts are run in tandem, and Claude synthesizes their outputs to reduce manual coordination and resolve conflicting results.

Do I need Codex and Gemini CLIs installed to coordinate parallel AI advisor prompts?

Yes, coordinating parallel AI advisor prompts requires both the Codex and Gemini CLIs to be installed locally, as the workflow relies on these command-line interfaces to execute and collect outputs.

What is the best way to consolidate conflicting results from multiple AI models?

The best way to consolidate conflicting results from multiple AI models is to orchestrate parallel prompts and apply synthesis, combining the different perspectives into a single, validated recommendation.

How does artifact generation work for synthesized AI recommendations?

Artifact generation for synthesized AI recommendations stores the coordinated outputs under the .omc/artifacts/ask/ directory, ensuring full traceability and auditability for architecture, design, and documentation tasks.