ask

Generate structured second opinions from local Claude or Gemini CLIs.

2|Updated May 11, 2026
One-click install
npx skills add https://github.com/xz1220/oh-my-kimi --skill ask-xz1220
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ask
Source: https://github.com/xz1220/oh-my-kimi/tree/main/skills/ask
Command: npx skills add https://github.com/xz1220/oh-my-kimi --skill ask-xz1220

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you get rapid, structured answers, reviews, or brainstorming from a locally installed advisor CLI without manually switching tools or losing reusable context.

Core Features & Use Cases

  • Backend selection for Claude/Gemini: Choose Claude/Anthropic or Gemini based on your intent and availability.
  • Reusable artifact output: Save the advisor conversation output into a timestamped markdown file for later reference.
  • Action-oriented results: Produce an artifact that includes the task, the exact prompt sent, the CLI’s raw output, a summary, and next steps.

Quick Start

Tell the skill what you want help with, and ensure you have the relevant local CLI installed (claude or gemini) so it runs locally and writes a markdown artifact under .omk/artifacts.

Frequently Asked Questions about ask

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

FAQPage Schema
How do I get a structured second opinion from a local Claude or Gemini CLI?

To get a structured second opinion from a local Claude or Gemini CLI, use this Skill to route your prompt to the installed backend, preserving the raw output and writing a markdown artifact to .omk/artifacts for reuse.

Can I save advisor CLI output as a markdown artifact for later reference?

Yes, you can save advisor CLI output as a markdown artifact. The Skill writes a timestamped markdown file containing the task, exact prompt, raw CLI output, summary, and next steps to the .omk/artifacts directory.

Do I need a specific local environment to run Claude or Gemini advisors for task review?

Yes, you need a local environment with the selected CLI backend installed. The Skill requires local execution by the chosen Claude or Gemini CLI to generate task review outputs and save them as artifacts.

What is the best way to use local advisors for brainstorming and iterative refinement?

The best way to use local advisors for brainstorming and iterative refinement is by selecting your desired backend, providing the task prompt, and letting the Skill execute it locally to produce an action-oriented markdown artifact.

Does this approach support both requirement clarification and critique generation?

Yes, this approach supports both requirement clarification and critique generation. It fits scenarios requiring structured outputs from local advisor CLIs, preserving the backend-specific prompt and raw output for productive review.

What are the limitations when switching between Claude and Gemini for prompt generation?

A limitation when switching between Claude and Gemini is that execution depends entirely on having the specific local CLI installed. The Skill preserves the backend-specific prompt but cannot run if the required CLI is missing.