VoltAgent/awesome-codex-subagents — reverse-engineered prompt

Build me a clean GitHub repo that is an awesome collection of Codex subagents. I want it to feel like a useful library people can clone, browse, and copy from when they want specialized AI helpers for coding work.

Create around 130 plus Codex subagent TOML files, grouped into clear categories like core development, language specialists, quality and security, DevOps, data, documentation, business, and other practical developer needs. Each agent should have a name, a plain description of when to use it, a model choice, reasoning effort, sandbox mode, and strong instructions for that role.

Write a polished README that explains what Codex subagents are, how to install them globally or inside one project, how precedence works, and that users need to explicitly ask Codex to delegate to them. Include copy paste terminal examples. Also add a simple contributing guide and keep the repo easy to scan and maintain.