skills.rest
cheafi
cheafiCOMMUNITY
Explore all skills by this author →
GitHub

ccg

Cross-validate tasks through Codex and Gemini and synthesize outputs.

Updated Feb 23, 2026
Search Tags:#code review#ai orchestration#conflict resolution#cross-validation#prompting#ux review#multi-model synthesis
One-click install
npx skills add https://github.com/cheafi/Trading-bot-CC --skill ccg-cheafi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ccg
Source: https://github.com/cheafi/Trading-bot-CC/tree/main/.github/skills/ccg
Command: npx skills add https://github.com/cheafi/Trading-bot-CC --skill ccg-cheafi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It reduces the risk of incorrect or one-sided outputs by routing a task through multiple AI models and reconciling their perspectives into a single result.

Core Features & Use Cases

  • Tri-model orchestration: runs Codex and Gemini in parallel, then synthesizes outputs into one unified answer.
  • Conflict-aware synthesis: explicitly surfaces disagreements and selects a final direction with rationale.
  • Practical fallbacks: continues with whichever provider is available, and drops to Claude-only synthesis if both are unavailable.

Example use cases include PR/code review that benefits from separate architecture vs. UX scrutiny and cross-validation for tasks where models may disagree on correctness or edge cases.

Quick Start

Give the AI the instruction: Review this change and cross-validate architecture/security with Codex while checking UX/clarity and edge cases with Gemini using ccg.

Frequently Asked Questions about ccg

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

FAQPage Schema
How do I cross-validate AI outputs to reduce hallucinations and bias?▼

Cross-validation reduces AI hallucinations by routing a single task through multiple models like Codex and Gemini, then synthesizing their outputs to surface disagreements and reconcile them into one unified answer.

What is the best way to run a code review that checks both architecture and UX clarity?▼

Multi-model code review separates architecture and security scrutiny from UX and edge-case checks by running Codex and Gemini in parallel, synthesizing their distinct perspectives into a single conflict-aware result.

How does multi-model synthesis work when AI providers disagree on correctness?▼

Multi-model synthesis explicitly detects conflicts when Codex and Gemini disagree on edge cases, then selects a final direction with rationale to reconcile the differing outputs into a unified answer.

Do I need parallel CLI invocations to cross-check answers across multiple models?▼

Parallel CLI invocations are required to run Codex and Gemini simultaneously for cross-checking answers, though the system applies provider fallback behavior to continue with whichever model is available.

What happens to cross-validation if Codex or Gemini is unavailable?▼

Provider fallback behavior ensures cross-validation continues with whichever provider is available, dropping to Claude-only synthesis if both Codex and Gemini are completely unavailable.

Can I use multi-model orchestration for architecture and UX deliverables?▼

Multi-model orchestration applies to architecture and UX deliverables by cross-validating the task across Codex and Gemini, explicitly surfacing disagreements to produce a reconciled final output.

Related Skills

Explore More Software Engineering Skills→
ClementATHClementATH

ccg

Route user requests through Codex and Gemini prompts and synthesize outputs.

Community
Advanced
dropsyoondropsyoon

ccg

Orchestrate parallel Codex and Gemini advisor runs and synthesize unified answers.

Community
Intermediate
BoraPerusicBoraPerusic

code-reviewer

Automate expert code review for multi-language codebases and pull requests.

Community
Advanced
DicklesworthstoneDicklesworthstone
70

cass

Index and search coding-agent sessions from multiple AI tools with JSON outputs.

Community
Advanced