Agent Skills: Command Architecture

>-

UncategorizedID: adaptive-enforcement-lab/claude-skills/command-architecture

Install this agent skill to your local

pnpm dlx add-skill https://github.com/adaptive-enforcement-lab/claude-skills/tree/HEAD/plugins/build/skills/command-architecture

Skill Files

Browse the full folder contents for command-architecture.

Download Skill

Loading file tree…

plugins/build/skills/command-architecture/SKILL.md

Skill Metadata

Name
command-architecture
Description
>-

Command Architecture

When to Use This Skill

A well-designed CLI has commands that work both independently and as part of larger workflows. This section covers:


Implementation

See the full implementation guide in the source documentation.

Key Principles

| Practice | Description | | ---------- | ------------- | | Flat hierarchy | Avoid deeply nested subcommands (max 2 levels) | | Verb-noun ordering | myctl restart deployment not myctl deployment restart | | Consistent flags | Use same flag names across commands | | Hidden internal commands | Mark debugging commands as hidden | | Exit codes | Use consistent exit codes (0=success, 1=failure, 2=usage error) |


Design commands for both humans and scripts.

Examples

See examples.md for code examples.

References