codex-support

Map canonical skill content to Claude and Codex target formats.

95|4|Updated Feb 27, 2026
One-click install
npx skills add https://github.com/aspenkit/aspens --skill codex-support
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-support
Source: https://github.com/aspenkit/aspens/tree/main/.agents/skills/codex-support
Command: npx skills add https://github.com/aspenkit/aspens --skill codex-support

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill coordinates multi-target outputs, mapping canonical skill content to target-specific formats for Claude and Codex, reducing duplication and keeping domains aligned.

Core Features & Use Cases

  • Centralized target and backend concepts that drive generation, routing, and content transforms across multiple targets.
  • Content transforms and references management to enable output for Claude, Codex, and future targets without rewriting skills.
  • Configurable persistence and per-target settings via a repo-wide .aspens.json, enabling safe, scalable publishing and collaboration.

Quick Start

Run doc-init to enable multi-target output routing for Claude and Codex.

Frequently Asked Questions about codex-support

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

FAQPage Schema
How do I output a single skill to both Claude and Codex without duplicating content?

To output a single skill to both Claude and Codex without duplication, map canonical skill content to target-specific formats using a multi-target output system. This orchestrates generation and routing across domains while keeping content aligned.

What does canonical-first generation mean for multi-target skill rendering?

Canonical-first generation for multi-target skill rendering means creating a single authoritative source content structure, then applying content transforms to render it for specific agents like Claude or Codex without rewriting the skill.

How do I configure per-target settings for multi-agent skill publishing?

Configure per-target settings for multi-agent skill publishing via a repo-wide .aspens.json file. This enables configurable persistence and safe, scalable publishing across Claude, Codex, and future targets.

Can I add future agent targets without rewriting existing skills?

Yes, you can add future agent targets without rewriting existing skills. Centralized target and backend concepts drive content transforms and references management, enabling output for new targets by mapping from the canonical content.

What is the best way to manage references and disk-fallback instructions across multiple agent backends?

The best way to manage references and disk-fallback instructions across multiple agent backends is using an orchestration system that applies per-target configuration. This satisfies content transforms and references without duplication across domains.

How do I enable multi-target output routing for Claude and Codex?

To enable multi-target output routing for Claude and Codex, run the doc-init command. This activates the system that orchestrates generation, routing, and content transforms across the configured targets.