conejo-code

Execute multi-stage coding workflows with adversarial model review and TDD.

Updated Jul 18, 2026

Search Tags:#typescript #workflow #tdd #coding #agent-dispatch #adversarial-review

Ready to run

npx skills add https://github.com/arthrod/conejo-skills --skill conejo-code

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: conejo-code
Source: https://github.com/arthrod/conejo-skills/tree/main/skills/conejo-code
Command: npx skills add https://github.com/arthrod/conejo-skills --skill conejo-code

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bun, vitest, madge, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill solves the problem of context loss and lack of adversarial rigor in AI-assisted coding by enforcing a strict, multi-stage workflow that prioritizes context preservation and adversarial review.

Core Features & Use Cases

Adversarial Coding Loop: Executes a rigorous feature lifecycle from brainstorming to implementation, ensuring every step is challenged by different AI models.
Context-Preserving Dispatch: Uses the Warren Protocol to delegate tasks to specialized agent CLIs or subagents, keeping the primary context window clean and focused.
Canonical Testing Doctrine: Enforces a strict red-green-refactor TDD workflow, ensuring no implementation occurs without a failing test.

Quick Start

Use the conejo-code skill to implement the new authentication service following the red-green TDD doctrine.

Frequently Asked Questions about conejo-code

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

FAQPage Schema

How do I maintain context and rigor when using AI agents for TDD coding workflows?▼

To prevent context loss in AI-assisted coding, use a workflow that delegates tasks to specialized agent CLIs or subagents. This context-preserving dispatch isolates complex subtasks, keeping the primary context window clean and focused on high-level orchestration.

What is the best way to enforce strict red-green-refactor TDD principles in backend logic development?▼

The best way to enforce red-green-refactor TDD principles in backend logic is to apply a canonical testing doctrine that prohibits implementation until a failing test exists. This ensures high reliability and strict adherence to test-driven development practices.

Do I need specific agent CLIs or subagent harnesses to execute adversarial model reviews?▼

Yes, executing adversarial model reviews requires a configured environment with access to agent CLIs or subagent harnesses. These tools facilitate the parallel task execution and adversarial validation necessary for the multi-stage coding loop.

Does this adversarial coding workflow support TypeScript and Bun dependencies?▼

Yes, the adversarial coding workflow supports TypeScript development and integrates with Bun, Vitest, and Madge dependencies. It applies directly to backend, logic, API, and CLI development tasks requiring high reliability and strict TDD adherence.

Why does my AI-assisted coding workflow lack adversarial validation during the implementation lifecycle?▼

AI-assisted coding lacks adversarial validation when it skips a rigorous feature lifecycle where every step is challenged by different AI models. Implementing an adversarial coding loop ensures each phase from brainstorming to implementation is thoroughly reviewed.