Z M Huang Claude Codex
z-m-huang-claude-codex is a Claude Code plugin for the Ship phase that orchestrates sequential AI reviews with a Codex final gate and hook-based step enforcement.
by Z-M-Huang · github.com/Z-M-Huang/claude-codex
Install this plugin when you want Claude Code sessions to chain Sonnet work with Codex review and hook-enforced gates before you ship.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install z-m-huang-claude-codex@Z-M-Huang/claude-codexBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:Z-M-Huang/claude-codex") and the paste-ready config with skillselion.get_install_config("plugin:Z-M-Huang/claude-codex").
What it does
z-m-huang-claude-codex is a Claude Code plugin that turns loose multi-model hacking into an orchestrated pipeline: work proceeds through sequential reviews, Codex acts as a final gate, and hooks enforce that you finish mandated steps before moving on. It targets solo and indie builders who already use Claude Code and sometimes lean on Codex for a second opinion but lack a repeatable ritual. You reach for it during active build work when agent tooling needs structure, and especially in the ship phase when every merge should pass a deliberate review chain rather than a single-pass skim. Compared with a plain agent skill, this is enforcement and routing at the Claude Code plugin layer—less about prompting tricks and more about making multi-AI collaboration the default path. Repository metadata lists about 25 GitHub stars and one plugin in the bundle; treat install popularity as unverified on Skillselion. If you only need occasional lint fixes, the overhead may not pay off; if you ship fast and regret review gaps, the pipeline framing is the point.
Highlights
- Sequential multi-AI review pipeline with Codex as the final approval gate
- Hook-based enforcement so task-based and multi-session flows cannot skip mandated steps
- Orchestrates Claude and Codex roles instead of manual model hopping in one session
- Task-based and multi-session patterns for longer solo-builder implementation arcs
- Single-plugin bundle from Z-M-Huang/claude-codex focused on pipeline discipline
Why builders use it
Solo builders switching between Claude and Codex ad hoc often skip structured review and have no hard stop before code ships.
After you register the plugin, Claude Code sessions can follow a enforced multi-AI review pipeline with Codex signing off at the end.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 25 stars, 0 votes.
FAQ
Who is z-m-huang-claude-codex for?
It is for solo and small-team developers using Claude Code who want Sonnet-to-Codex pipelines and enforced review gates instead of informal model switching.
When should I use z-m-huang-claude-codex?
Use it when you are implementing features or preparing merges and need sequential AI reviews with a Codex final gate before you call the work done.
How do I add z-m-huang-claude-codex to my agent?
Add the Z-M-Huang/claude-codex marketplace plugin to Claude Code from the repo bundle, then enable hooks and follow the task-based or multi-session workflow described in the plugin docs.
Comments
Share how you use z-m-huang-claude-codex, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.