/convert-codex — Repair Codex Parity for an HQ Root

SkillAI & models

Add Codex support to older Claude-first HQ roots without breaking Claude setup.

Available today. Use it from your connected AI after setup.

No other account needed.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the /convert-codex — Repair Codex Parity for an HQ Root skill

What this skill tells your AI

The instructions your AI receives, as published by indigoai-us/hq-core in .claude/skills/convert-codex/SKILL.md and read by ahel’s review.

Add Codex-facing files and bridges to an existing HQ that was originally built around Claude Code.

Args: $ARGUMENTS

Run a dry-run unless the user explicitly passes --apply:

bash core/scripts/convert-codex.sh $ARGUMENTS

AGENTS.md when missing.
.codex/config.toml when missing.
.codex/claude bridge to .claude/.
.codex/output-style.md bridge to the active .claude/output-styles/ file.
.agents/skills exposure for Codex skills (replaces the legacy .codex/prompts → .claude/commands/ bridge — commands tree is gone).
Missing agents/openai.yaml metadata for existing skills.

Do not overwrite existing files.
Do not replace an existing .agents/skills directory with a symlink.
If .agents/skills already exists as a real directory, only add missing skill folders.
If a requested repair requires changing existing content, stop and ask before proceeding.

Preview:

bash core/scripts/convert-codex.sh --dry-run

Repair current root:

bash core/scripts/convert-codex.sh --apply

Repair another HQ root:

bash core/scripts/convert-codex.sh --root=/path/to/hq --apply

Advanced