dev-ccg

Name: dev-ccg
Availability: InStock
Author: EvolutionAPI

Run the same prompt through Claude, Codex, and Gemini in parallel and synthesize results.

520|175|Updated Apr 8, 2026

Search Tags:#code-review #risk-mitigation #architecture-decision #multimodel #llm-orchestration #tri-model #prompt-synthesis

Ready to run

npx skills add https://github.com/EvolutionAPI/evo-nexus --skill dev-ccg

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dev-ccg
Source: https://github.com/EvolutionAPI/evo-nexus/tree/main/.claude/skills/dev-ccg
Command: npx skills add https://github.com/EvolutionAPI/evo-nexus --skill dev-ccg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Tri-model orchestration addresses the risk of single-model bias by evaluating the same prompt across Claude, Codex, and Gemini in parallel and selecting the most robust answer.

Core Features & Use Cases

Parallel evaluation: Run the same prompt in Claude, Codex, and Gemini simultaneously.
Synthesis & validation: Aggregate and compare responses to highlight agreement and discrepancies; synthesize a final answer.
Use Case: For high-stakes architectural decisions or critical code reviews where model consensus improves reliability.

Quick Start

Submit a single prompt topic to be run in parallel across Claude, Codex, and Gemini and retrieve a synthesized best answer.

Frequently Asked Questions about dev-ccg

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

FAQPage Schema

What is tri-model consensus for code review and architecture decisions?▼

Parallel model evaluation runs your prompt simultaneously through Claude, Codex, and Gemini to aggregate responses and synthesize a final answer. It highlights agreement and discrepancies, significantly reducing single-model bias for critical architectural decisions and complex analyses.

How do I run a prompt across Claude, Codex, and Gemini simultaneously?▼

Submit a single prompt topic to trigger parallel evaluation across Claude, Codex, and Gemini. The orchestration engine aggregates the responses, highlights discrepancies, and synthesizes a final validated answer saved to a structured workspace.

Do I need API access to Codex and Gemini to use tri-model orchestration?▼

External API access to Codex and Gemini is optional. The orchestration engine handles fallback to single-model mode if those APIs are unavailable, ensuring prompt synthesis still executes and saves results to the structured workspace.

When should I use multi-model LLM orchestration for risk mitigation?▼

Use multi-model orchestration for high-stakes architectural decisions, critical code reviews, and complex analyses. It applies tri-model prompt synthesis to mitigate risk, improving reliability when single-model output lacks sufficient consensus for risk mitigation.

What are the limitations of tri-model consensus when APIs are unavailable?▼

The primary limitation occurs when external APIs are unavailable, triggering fallback to single-model mode. Without full API access, parallel evaluation stops, preventing cross-model discrepancy comparison and reducing the synthesized result's reliability.