skills.rest
MathieuDoyon
MathieuDoyonCOMMUNITY
Explore all skills by this author →
GitHub

mission-codex

Delegate TDD and debug sub-missions to the OpenAI Codex CLI and return a structured mission-report.

Updated May 22, 2026
Search Tags:#tdd#debugging#ai orchestration#codex cli#code delegation#git diff verification#cost-tier routing
One-click install
npx skills add https://github.com/MathieuDoyon/mission --skill mission-codex
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mission-codex
Source: https://github.com/MathieuDoyon/mission/tree/main/.claude/skills/mission-codex
Command: npx skills add https://github.com/MathieuDoyon/mission --skill mission-codex

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It reduces mission iteration time and cost by delegating low-risk coding tasks to Codex, while still producing structured outcomes the orchestrator can act on when work succeeds or blocks.

Core Features & Use Cases

  • Cost-tier delegation wrapper: Routes mission-tdd and mission-debug work to the OpenAI Codex CLI when configured and available.
  • Deterministic pre-flight guard: Verifies Codex availability via agents check codex and immediately reports blocked if unusable.
  • Outcome parsing with escalation signals: Converts Codex stdout/stderr and diff evidence into a mission-report the orchestrator uses to continue, retry, or escalate to the primary Claude sub-skill.

Quick Start

Ask Claude to run the codex delegation wrapper to complete the currently selected cheap-tier TDD or debug sub-mission for your Mission objective.

Frequently Asked Questions about mission-codex

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

FAQPage Schema
How do I delegate test-driven development tasks to OpenAI Codex CLI safely?▼

You can delegate test-driven development tasks by routing cheap-tier sub-missions to the OpenAI Codex CLI wrapper, which executes synchronously and returns a structured mission-report with git diff verification for safe code changes.

Can I use Codex for AI orchestration without manually verifying code changes?▼

Yes, AI orchestration with Codex automatically captures stdout and stderr, then applies git diff-based verification to validate code changes and return explicit status mappings like complete, blocked, or needs more context.

What happens if Codex CLI is unavailable during mission orchestration?▼

If Codex CLI is unavailable, a deterministic pre-flight guard checks availability and immediately reports a blocked status, preventing execution and allowing escalation to the primary sub-skill.

Does cost-tier routing for debugging tasks require pre-flight checks?▼

Yes, cost-tier routing for debugging requires Codex pre-flight checks to verify availability before execution, ensuring reliable delegation and structured outcome parsing for mission escalation signals.

When should I avoid using Codex delegation for TDD sub-missions?▼

You should avoid Codex delegation when the orchestrator needs direct reasoning about complex code changes, as this wrapper is designed for fast, cost-controlled, low-risk tasks rather than primary logic implementation.

Related Skills

Explore More Software Engineering Skills→
MathieuDoyonMathieuDoyon

mission-opencode

Delegate cheap-tier TDD and debug tasks to an OpenCode CLI agent with structured mission reporting.

Community
Intermediate
MathieuDoyonMathieuDoyon

mission-pi

Delegate cheap-tier mission tasks to the Pi agent with synchronous commands and structured reports.

Community
Intermediate
kimprobablykimprobably

spec-kitty-mission-system

Explain Spec Kitty mission types, phases, artifacts, and guards for workflow advancement.

Community
Advanced
alexiolanalexiolan
17

craft-duo

Delegate data-layer tasks to Codex while Claude handles architecture and UI.

Community
Advanced