ask

Route prompts to local Claude, Codex, or Gemini CLIs and store responses as artifacts.

Updated May 17, 2026
One-click install
npx skills add https://github.com/tiankong0101-byte/skills-registry --skill ask-tiankong0101-byte
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ask
Source: https://github.com/tiankong0101-byte/skills-registry/tree/main/skills/oh-my-claudecode/skills/ask
Command: npx skills add https://github.com/tiankong0101-byte/skills-registry --skill ask-tiankong0101-byte

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of manually choosing and invoking the correct local AI CLI by providing a single, standardized routing path for Claude, Codex, or Gemini.

Core Features & Use Cases

  • Advisor Routing: Sends a prompt through the appropriate local model interface based on the selected provider.
  • Artifact Capture: Persists each run as an ask artifact for later review and traceability.
  • Safe Invocation Guardrails: Prevents manual raw CLI assembly so the wrapper controls flags, compatibility, and persistence.
  • Use Case: Use it to ask a coding model to review a patch, draft an implementation plan, or suggest product improvements without switching tools.

Quick Start

Use the ask skill to route my prompt to codex and save the result as an artifact.

Frequently Asked Questions about ask

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I route a single prompt to different local AI CLIs like Claude or Codex?

You can route a prompt to Claude, Codex, or Gemini using a unified wrapper that standardizes AI CLI invocation. The wrapper automatically controls command-line flags and compatibility while selecting the appropriate local provider for your task.

Can I save CLI automation responses as artifacts for later review?

Yes, artifact capture persists each AI CLI run as an ask artifact under a local directory. This provides traceability and allows you to review the stored responses from Claude, Codex, or Gemini at a later time.

Do I need to manually assemble command-line flags when using multiple AI providers?

No, safe invocation guardrails prevent manual raw CLI assembly by having the wrapper control flags, compatibility, and persistence. This removes the friction of manually choosing and invoking the correct local AI CLI.

What is the best way to use local AI models for code review or planning without switching tools?

Use a standardized routing path to ask a coding model to review a patch, draft an implementation plan, or suggest product improvements. It handles provider selection automatically so you can perform ideation tasks without switching tools.

What prerequisites are required to execute AI routing for local provider CLIs?

Executing AI routing requires omc ask execution, installed authenticated provider CLIs, and artifact persistence configured under the local ask directory. You must have these local provider interfaces set up beforehand.

Are there limitations when using a unified wrapper for local AI CLI automation?

The wrapper is limited to routing to installed authenticated local provider CLIs and does not support manual raw CLI assembly. It applies specifically to review, planning, and ideation tasks requiring provider selection.