codex-delegation

Route development tasks between Bravo and Codex agents for parallel execution.

1|Updated Apr 18, 2026
One-click install
npx skills add https://github.com/CC90210/CMO-Agent --skill codex-delegation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-delegation
Source: https://github.com/CC90210/CMO-Agent/tree/main/skills/codex-delegation
Command: npx skills add https://github.com/CC90210/CMO-Agent --skill codex-delegation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @openai/codex, and includes scripts (resource) components.

What problem does it solve?

This skill solves the coordination challenge between multiple AI agents by intelligently routing tasks to the most capable engine, preventing context loss and ensuring optimal parallel execution.

Core Features & Use Cases

  • Dual-AI Routing: Automatically delegates backend, debugging, and adversarial review tasks to Codex while keeping creative and strategic work within Bravo.
  • Parallel Execution: Enables simultaneous task processing, such as having one agent implement a feature while another performs a background code review.
  • Failure Recovery: Implements a robust 3-strike rule and pre-flight checks to handle CLI staleness, authentication issues, and task timeouts gracefully.

Quick Start

Delegate the backend investigation of the current API timeout to the Codex agent using the background execution mode.

Frequently Asked Questions about codex-delegation

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

FAQPage Schema
How do I handle CLI staleness and authentication issues during AI agent delegation?

Yes, you can delegate backend debugging to Codex in background execution mode while Bravo handles creative feature implementation simultaneously. This parallel execution allows one agent to investigate API timeouts while another performs systematic code reviews.

Do I need the Codex CLI to route full-stack feature implementation tasks?

You need the Codex CLI installed to use dual-AI task routing for backend debugging and adversarial code reviews. The Codex CLI serves as the required dependency for delegating complex development tasks away from the primary Bravo agent.

What is the best way to delegate adversarial code reviews to a secondary AI agent?

Dual-AI routing delegates backend, debugging, and adversarial review tasks to Codex while keeping creative and strategic work within Bravo. This division maximizes parallel execution and ensures optimal task routing for full-stack features.

When should I use dual-AI routing for backend debugging instead of a single agent?

Dual-AI routing specifically targets full-stack feature implementation, adversarial code reviews, and systematic backend debugging. It requires specific context injection protocols to ensure agent alignment and task success across these development scenarios.

How does multi-agent task routing prevent context loss during parallel execution?

Multi-agent task routing prevents context loss by intelligently delegating complex development tasks between Bravo and Codex engines, ensuring each agent operates within its most capable domain while maintaining architectural integrity. It applies context injection protocols to align agents.

How do I handle CLI staleness and authentication issues during AI agent delegation?

AI agent delegation handles CLI staleness and authentication issues through robust pre-flight checks and a 3-strike failure recovery rule. This mechanism gracefully manages task timeouts and ensures agent alignment before executing complex routing.