A² · Agentic AI × Anything：多 agent 编排方法论与治理 skill 合集（旗舰 cto-orchestration）
Skill Claude CodeCodex
A handbook for building and reviewing back-end software that uses AI agents or language models, stores data in databases, or needs strict repository checks. It covers design, prompts, tools, memory, search, reliability, security, testing, cost, and data access.
Skill Claude CodeCodex
A set of engineering guidelines for building consistent front-end interfaces with AI agents. It covers shared design components, type-checked interfaces, visual checks and staged testing requirements.
Skill Claude CodeCodex
An asynchronous mailbox system for communication between long-running or independently managed software agents. Each agent has its own inbox, and messages can be sent, read, replied to, or archived.
Skill Claude CodeCodex
A workflow for coordinating several coding agents on software delivery. It assigns implementation, review, and operational work through a shared control process with goals, status checks, and handoffs.
Skill Claude CodeCodex
A set of rules for collaborating on Git repositories, where Git records code changes and branches let people work separately. It describes when to use feature branches, pull requests, rebasing, reviews, and protected branches.
Skill Claude CodeCodex
A standard for making backend software observable, meaning easier to understand when something goes wrong. It connects logs, traces of requests across services, and business identifiers using OpenTelemetry, a common monitoring standard.
Skill Claude CodeCodex
A reusable method for coordinating multiple AI agents across different kinds of work. It explains how to split tasks, pass clear instructions, combine results and check them with independent reviews.
Skill Claude CodeCodex
A one-time setup guide for basic project governance in a new software repository. It creates a small documentation structure for decisions, current work, modules, and a roadmap, without introducing a large project-management system.
Skill Claude CodeCodex
Use when a task involves reading, editing, reviewing, testing, debugging, refactoring, or otherwise working with source code files. Source paths include /.{py,ts,tsx,js,jsx,mjs,go,rs,java,kt,swift,rb,c,cc,cpp,h,hpp,sh,bash,zsh,sql,vue,svelte}. Do not use for docs-only, planning-only, or non-code file tasks unless the…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: