skills.rest
hotaq
hotaqCOMMUNITY
Explore all skills by this author →
GitHub

ask-claude

Routes a focused question or review to the local Claude CLI and captures the answer as a markdown artifact.

Updated Apr 20, 2026
Search Tags:#code review#cli#claude#second opinion#local execution#artifact capture
One-click install
npx skills add https://github.com/hotaq/Sprite_harmess --skill ask-claude-hotaq
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ask-claude
Source: https://github.com/hotaq/Sprite_harmess/tree/main/.codex/skills/ask-claude
Command: npx skills add https://github.com/hotaq/Sprite_harmess --skill ask-claude-hotaq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill lets you send a focused question or review to the local Claude CLI when you want a fast second opinion without switching to remote routing.

Core Features & Use Cases

  • Local advisory workflow: Runs Claude through the canonical local command path so the interaction stays on your machine.
  • Reusable artifact capture: Saves the prompt, output, summary, and next steps into a markdown artifact for later reference.
  • Practical uses: Great for code reviews, design feedback, research questions, and short implementation checks that benefit from an external perspective.

Quick Start

Ask the skill to run your question through the local Claude CLI and save the result as a reusable artifact.

Frequently Asked Questions about ask-claude

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

FAQPage Schema
How do I get a local code review from Claude without using remote routing?▼

You can run a focused code review through the local Claude CLI to get a second opinion. This routes your question through the canonical local command path on your machine and captures the output as a reusable markdown artifact.

Can I save Claude CLI output as a markdown file for later reference?▼

Yes, saving Claude CLI output as a markdown file captures the original prompt, raw output, summary, and next steps. It writes these markdown artifacts under a timestamped slug in the .omx/artifacts directory for later reference.

Do I need a specific binary installed to run local Claude advisory tasks?▼

Yes, you need a working claude binary installed locally. The Skill requires this local executable to route questions through the canonical OMX CLI path instead of relying on remote routing.

What is the best way to capture a second opinion from Claude for code reviews?▼

The best way to capture a second opinion is routing your review request to the local Claude CLI. This approach runs locally on your machine and saves the prompt, output, summary, and next steps into a markdown artifact.

How does local CLI execution for AI advisory tasks work?▼

Local CLI execution for AI advisory tasks works by sending a focused question to the local Claude binary through the canonical OMX path. The interaction stays entirely on your machine and writes the captured response to a timestamped markdown file.

What are the limitations of running Claude locally for code reviews?▼

A key limitation is the dependency on a working local claude binary. It is designed for short advisory tasks and focused questions rather than extensive remote processing workflows.

Related Skills

Explore More Software Engineering Skills→
riethmayerriethmayer
2

ask-claude

Query Claude locally via the OMX CLI and save artifacts to .omx/artifacts.

Community
Intermediate
OrinksOrinks
24

ask-gemini

Execute the local Gemini CLI to generate markdown feedback artifacts.

Community
Intermediate
jchildreejchildree
1

faceless-review

Generate concise structured code review comments for PRs.

Community
Intermediate
chrisxue90chrisxue90
2

ask-claude

Route queries through the local Claude CLI and save artifacts.

Community
Intermediate