ccg

Merge Codex and Gemini advisor outputs into one conflict-aware answer.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/nichobbs/lyric-lang --skill ccg-nichobbs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ccg
Source: https://github.com/nichobbs/lyric-lang/tree/main/.claude/skills/ccg
Command: npx skills add https://github.com/nichobbs/lyric-lang --skill ccg-nichobbs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

CCG reduces the risk of one-perspective answers by orchestrating parallel Codex and Gemini advisor runs, then synthesizing them into a single, coherent response.

Core Features & Use Cases

  • Parallel advisor perspectives: runs Codex-focused guidance and Gemini-focused UX/design guidance for broader coverage.
  • Conflict-aware synthesis: reconciles disagreements by explicitly surfacing conflicts and choosing a final direction with rationale.
  • Artifact-based integration: gathers outputs from .omc/artifacts/ask/ and combines the latest Codex and Gemini results.

Quick Start

Use the ccg skill to run both advisor prompts and synthesize the results by asking: Review this request for architecture and backend risks, then review it again for UX, readability, and alternative approaches.

Frequently Asked Questions about ccg

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

FAQPage Schema
How do I synthesize multiple advisor outputs for architecture and UX design?

To synthesize multiple advisor outputs for architecture and UX design, run parallel Codex and Gemini advisor prompts. The system reconciles disagreements by surfacing conflicts and choosing a final direction with rationale.

What is the best way to reconcile conflicting code review perspectives from different AI models?

Reconciling conflicting code review perspectives involves orchestrating parallel AI advisor runs and applying conflict-aware synthesis. This approach merges results while explicitly highlighting disagreements and providing a chosen final direction.

How do I combine Codex and Gemini responses into a single coherent answer?

Combining Codex and Gemini responses requires invoking both advisor commands and reading their latest artifacts. The system then merges these outputs into one unified response that addresses architecture, backend, and documentation requests.

Can I use this orchestration approach for both backend architecture and documentation requests?

Yes, this orchestration approach applies to architecture, backend, UX design, and documentation requests. It gathers outputs from both advisors to provide broader coverage and a merged response for complex tasks.

How does artifact-based integration work when merging AI advisor outputs?

Artifact-based integration works by gathering outputs stored under the artifacts ask directory. The system reads the latest Codex and Gemini results from these files and combines them into a single synthesized response.

When do I need parallel AI advisor perspectives for complex tasks?

You need parallel AI advisor perspectives when complex tasks require multiple viewpoints to reduce the risk of one-perspective answers. This is crucial when potential disagreements must be reconciled for architecture or UX design decisions.