noobnooc agent skills
Agent
Two skills for coding agents: wtf audits a project before shipping, better-codex keeps the agent disciplined while working on code
- 1,387 ★ on the repository
- Last commit
Extended access
Why this level
- Directs the agent to read project files and run check commands: git, lint, tests, build
- better-codex governs how the agent edits project files
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/agentDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Directs the agent to read project files and run check commands: git, lint, tests, build
- better-codex governs how the agent edits project files
This is third-party code. Review the repository files before installing.
What it does
The repository keeps two portable skills in folders under skills/. The wtf skill runs a strict audit before a commit or deploy: it checks project structure, traces of secrets and env vars, cache and build artifacts tracked in git, migrations and schema drift, dead code and the usual mistakes of quickly built apps, then reports findings by severity from blocker to minor. The better-codex skill is a set of behavior rules for codebase work: keep to the stated scope, re-read a file before editing, do not overwrite other people's changes, find the root cause of a bug, pick dependencies deliberately, and keep internal reasoning out of code and UI. Each skill is a folder with a SKILL.md and Codex metadata under agents/. You copy the skills into your agent's skills directory and call them by name.
Who it is for. For developers who work with coding agents and want predictable behavior and a pre-release project check.
Good fit when
- Before a commit or deploy you need to find blockers: secrets, migrations, dead code, deploy risks
- You want the agent to stay within scope and not overwrite other people's edits
- You need a root-cause fix rather than a patch over the first suspicious symptom
Not a fit when
- You need one-off advice rather than standing behavior rules for the agent
- Your agent does not read skills from a folder and does not support the SKILL.md format
Example request
Audit this project with wtf before deploy and show the blockersLimitations
This is a set of instructions, not a standalone program: the result depends on how closely the agent follows the rules. The agents/ metadata targets Codex; in other agents the SKILL.md itself does the work. The wtf audit relies on commands and tools available in the agent's environment.
How to disable. Remove the wtf and better-codex folders from your agent's skills directory, for example ~/.codex/skills/.
FAQ
How does wtf differ from better-codex?
wtf is a one-time pre-release audit that reports findings by severity. better-codex is a standing set of behavior rules the agent follows while working on code.
Does it work outside Codex?
Yes. The SKILL.md files use a universal format and suit any agent that reads skills from a folder. The agents/ folder adds Codex-specific metadata.
Discussion
No questions or reviews yet.
Related
SkillsA library of 30-plus instruction skills for coding agents: planning, building, testing, hallucination guards and shipping
61 small skills for Codex across 13 areas: codebase orientation, tests, migrations, security, CI and pull request review
A finalization-discipline skill: the agent describes the accepted result and keeps rejected alternatives out of titles, commits and PRs
A skills library for Claude Code, Cursor and Codex: team code review, TDD, stacked PRs and repository readiness scoring
Two skills for Codex and Claude Code: prd keeps a living implementation checklist and review audits designs from the user's side
A set of independent skills against drift in large codebases: hard-cut, root-cause tracing, lean testing and CI right-sizing.
README in short
The README presents the repository as a personal set of agent skills where each skill is a portable folder with a SKILL.md and optional agents/ metadata. A table describes two skills: wtf for a pre-launch and pre-commit project audit, and better-codex as a reliability overlay for codebase work. Install is via git clone plus copying or symlinking the folders into the agent's skills directory, for example ~/.codex/skills/. It shows how to call a skill by name and how to maintain skills when they change.
README badge
Original SKILL.md
--- name: better-codex description: Behavioral guardrails for Codex coding work based on common user complaints. Use when Codex is asked to implement, modify, debug, review, test, or operate on a codebase and should avoid unsafe scope expansion, stale edits, fake completion, brittle edits, shallow debugging, superficial patch-on fixes, one-off special-case code, short-term design choices that hurt maintainability, blindly following incorrect user assumptions, poor dependency choices, generic product or UI output, over-mocked tests, noisy approvals, verbose status reports, or leaking internal reasoning into project artifacts or user-facing UI. --- # Better Codex Use these rules as a reliability overlay for codebase work. They convert recurring public complaints about Codex into prompt-fixable behavior. Do not cite the complaints during normal task execution; just follow the rules. ## Core Contract - Treat the user's request as the full scope. Stay inside the current workspace and the paths the user named unless the user explicitly authorizes a broader scope. - Treat investigation, review, audit, diagnosis, or planning as read-only unless the user also asked for implementation. - Do not mutate global configuration, authentication files, Git remotes, credentials, files outside the workspace, generated lockfiles, or many repositories at once without explicit user authorization. - When the task requires a risky or broad mutation, first state the exact files, commands, and expected effect, then wait for confirmation.