clavain

Orchestrate the software development lifecycle using Claude, Codex, and GPT-5.2 Pro.

3|1|Updated Feb 15, 2026

Search Tags:#code review #code generation #workflow orchestration #ai agent #development lifecycle #multi-model

Ready to run

npx skills add https://github.com/mistakeknot/Demarch --skill clavain

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: clavain
Source: https://github.com/mistakeknot/Demarch/tree/main/.gemini/generated-skills/clavain
Command: npx skills add https://github.com/mistakeknot/Demarch --skill clavain

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the entire software development lifecycle, from initial brainstorming to shipping code, by orchestrating multiple AI models and managing complex workflows.

Core Features & Use Cases

Full Lifecycle Orchestration: Manages phases from discovery and brainstorming to execution, review, and shipping.
Multi-Model Coordination: Integrates Claude, Codex, and GPT-5.2 Pro for diverse task execution.
Use Case: Kickstart a new project by having Clavain onboard your project, setting up documentation, and generating an initial roadmap, then use its routing to guide the AI through coding and review phases.

Quick Start

Use the clavain skill to onboard a new project and set up its development environment.

Frequently Asked Questions about clavain

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

FAQPage Schema

How do I orchestrate the full software development lifecycle using multiple AI models?▼

You can orchestrate the full software development lifecycle by routing tasks through heterogeneous AI models. This Skill coordinates Claude, Codex, and GPT-5.2 Pro to manage phases from discovery and brainstorming to execution, review, and shipping.

What is the best way to coordinate code generation and code review across different AI agents?▼

The best way to coordinate code generation and review across AI agents is using a layered architecture with skills, agents, commands, and hooks. This setup routes specific development tasks to the appropriate AI model for execution and automated review.

Can I use workflow orchestration to manage my project from initial planning to shipped code?▼

Yes, you can use workflow orchestration to manage your project from planning to shipped code. The Skill onboards your project, generates an initial roadmap, sets up documentation, and guides the AI through coding and review phases.

Does this multi-model AI agent approach support project discovery and strategy planning phases?▼

Yes, this multi-model AI agent approach supports project discovery and strategy planning phases. It explicitly manages these initial stages before moving into execution, ensuring your development lifecycle starts with structured problem discovery and strategic alignment.

Why do I need multi-model coordination instead of a single AI model for my development workflow?▼

You need multi-model coordination to leverage the distinct strengths of different AI models within your development workflow. Integrating Claude, Codex, and GPT-5.2 Pro allows the system to route diverse lifecycle tasks to specialized models for better execution.

What are the limitations of using heterogeneous AI models for workflow orchestration?▼

A limitation of using heterogeneous AI models for workflow orchestration is the dependency on the availability and integration of external models like Codex and GPT-5.2 Pro. You must ensure your environment supports these connections for the layered architecture to function.