AnastasiyaWCommunity·27 Agent Skills Included

codex-claude-code-config

Battle-tested rules, hooks, and skills for coding agents

Equips coding agents with proven architectural principles, safety hooks, and domain skills for long-running projects. Stops agents from claiming unfinished work, leaking secrets, over-engineering code, or losing context across sessions. Includes verification loops, code review workflows, frontend and ML skills, and ready-to-install guardrails for Claude Code and Codex.

140|23

↗

npx skills add AnastasiyaW/codex-claude-code-config --all -g -y

Available:

Tells the agent how to pick and apply the right principles, hooks, and skills for a project, enforces a stop-when-sufficient reasoning policy, and defines always-on working rules like no-guessing, finish-the-task, and verification before claiming done.

All Skills in This Repository (27)

Pure Emerald Level Indicators

📦 In Repo

AnastasiyaW

lean-code

Enforces minimal coding by prioritizing standard library and native platform features.

agent-harness-design

Design agent harnesses with risk taxonomies, permission decision objects, and structured event models.

plan-to-tickets

Decompose project plans into Markdown tickets with acceptance criteria.

proof-verify

Enforces a plan-based verification workflow using frozen specifications and independent agents.

distill-feedback

Extract durable user corrections from JSONL feedback logs into atomic configuration rules.

repo-map

Generates ranked lists of structurally important symbols in a codebase using PageRank algorithms.

workflow-orchestration

Orchestrate deterministic multi-agent workflows with a JavaScript-based runtime.

feature-new

Scaffold new feature documents with F-NNN identifiers and update tracking files.

layer-new

Scaffold bounded concern layers with directories, history logs, and principle placeholders.

ml-research-lab

Standardize machine learning research workflows with verification gates for data quality and production readiness.

notebooklm-grounded-research

Retrieve citation-backed answers from document corpora via the NotebookLM MCP bridge.

pixel-art-storyboard

Convert narrative synopses into self-contained HTML5 canvas animations.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install codex-claude-code-config?▼

Run `npx skills add AnastasiyaW/codex-claude-code-config --all -g -y` in your terminal to install all skills globally.

What does this configuration system do?▼

It gives your coding agent battle-tested working rules, safety hooks, and domain skills so it finishes tasks, verifies its own output, and avoids common failure modes like over-engineering or context loss.

Does it work with both Claude Code and Codex?▼

Yes. The AGENTS.md and CLAUDE.md files are designed for cross-harness use, and the skills follow the standard SKILL.md format that works in Claude Code, Codex, and other compatible agents.

How do the safety hooks protect my project?▼

The hooks block destructive commands, detect leaked API keys, enforce deletion confirmation, and validate that claimed work is actually verified before a session ends.

Can I install only part of this configuration?▼

Yes. The system is modular: you pick the principles, hooks, and skills that match your project type instead of copying everything.