Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Designs CLI surface area before or during implementation: command trees, flags, subcommands, help text, output formats, exit codes, and config precedence. Applies the clig.dev rubric to produce a complete, implementable spec rather than ad-hoc interface decisions.
Producing a full spec first prevents interface changes after users depend on the CLI, which are expensive to backtrack.
When to use it
- Spec out a new build tool's subcommands and flags before writing any code
- Audit an existing CLI for inconsistent flag naming across subcommands
- Define exit codes and error messages for a deployment script
- Add --json output and --dry-run support to a command that lacks them
- Establish config file vs env var precedence for a multi-environment tool
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: design a CLI for a file sync tool that targets both humans and scripts
- 1Asks minimum clarifying questions: command name, input sources, interactivity needs, and output contract
- 2Produces a command tree with init, sync, and status subcommands
- 3Defines a flags table with types, defaults, and required/optional markers
- 4Specifies stdout vs stderr contract, --json mode, and exit code map
- 5Outputs 5-10 example invocations including piped and --no-input cases
A compact CLI spec the user can hand directly to an implementer or use as a coding prompt