awirectl

name	awirectl
description	Use when operating the awirectl generated CLI. Discover commands, inspect parameters, check auth state, and execute API operations safely.

awirectl CLI

Use this skill when a user asks you to operate awirectl, inspect its API commands, or find the right generated command for an API task.

Workflow

1. Search for candidates with awirectl search "<intent>" --json; use --limit when needed. Search is only candidate discovery.
2. Inspect the exact command with awirectl commands show <path...> --json before executing an unfamiliar command.
3. If the command detail has auth.required=true, run awirectl auth status --hostname <host> before execution. Use http.default_hostname when present unless the user provides --hostname or $AWIRECTL_HOST.
4. Execute only after flags, body, auth, HTTP path, and output hints are clear from commands show.

General Commands

• awirectl commands --json: full generated command catalog.
• awirectl commands --include-hidden --json: include hidden generated commands.
• awirectl commands show <path...> --json: source of truth for one command.
• awirectl commands schema --json: catalog schema version for parser compatibility.
• awirectl search "<intent>" --json: ranked candidate commands.

Maintenance Commands

• awirectl --version or awirectl -v: print CLI build version.
• awirectl update: update this CLI from configured GitHub Releases. Run only when the user explicitly asks to update awirectl; it may replace the current executable. Use --yes only when explicitly authorized.

References

• Read references/catalog.md for the command discovery protocol and catalog field meanings.
• Read references/modules/awire.md for the awire module command index.

Rules

• Do not guess flags or request body shape from command names.
• Do not execute directly from search results; confirm with commands show first.
• Prefer -o json for machine-readable command output unless the user asks for human-readable output.
• Use --file, --set, or --set-str for JSON request bodies according to commands show body requirements.
• For sensitive flags, prefer safe modes from flags[].input_modes: --<flag>-env, --<flag>-file, or --<flag>-stdin.