This skill provides a reliability overlay for Codex when implementing, debugging, or reviewing code. It is designed for AI agent power users who need to prevent brittle edits, shallow debugging, and the tendency of models to follow incorrect user assumptions.
Follow the documented setup, then try a first task.
Installs: Complete documented repository bundle: wtf and better-codex · Codex
Run from: Directory where the agent repository should be cloned
In a terminal, clone the repository, enter its root directory, create the Codex skills directory, and copy both skill folders into it.
git clone https://github.com/noobnooc/agent.git cd agent mkdir -p ~/.codex/skills cp -R skills/wtf skills/better-codex ~/.codex/skills/
Installs: Complete documented repository bundle: wtf and better-codex, linked for local development · Codex
Run from: Root of the local repository checkout
In a terminal, from the local repository root, create the Codex skills directory and symlink both skill folders into it.
mkdir -p ~/.codex/skills ln -s "$(pwd)/skills/wtf" ~/.codex/skills/wtf ln -s "$(pwd)/skills/better-codex" ~/.codex/skills/better-codex
Uses wtf
Use $wtf to audit this project before I commit or deploy it.
Uses better-codex
Use $better-codex to complete this coding task safely and concisely.
Claude
Codex
Antigravity