ask

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

3|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ask-marcmunoz-uno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ask
Source: https://github.com/marcmunoz-uno/jailbreak/tree/main/skills/ask
Command: npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ask-marcmunoz-uno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of needing a consistent way to send prompts to local Claude, Codex, or Gemini CLIs while ensuring the results are captured and saved for later use.

Core Features & Use Cases

  • Provider-agnostic advisor routing: Send the same task to Claude, Codex, or Gemini through a single canonical flow.
  • OMC-managed execution: Use the omc ask wrapper for correct provider selection, compatibility, and artifact persistence.
  • Ask artifact capture: Persist each routed response as an ask artifact for auditing, follow-up, and reuse; for example, you can route a security review question for a patch and keep the saved output as a durable record.

Quick Start

Use the command /oh-my-claudecode:ask codex "review this patch from a security perspective".

Frequently Asked Questions about ask

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

FAQPage Schema
How do I route prompts to local Claude, Codex, or Gemini CLIs consistently?

Use the omc ask wrapper to route prompts to locally installed Claude, Codex, or Gemini. It enforces a canonical execution path for correct provider selection, version-safe flag handling, and consistent artifact persistence.

Can I save Claude or Codex CLI responses as durable artifacts for auditing?

Yes, each routed response is captured as an ask artifact and written to the .omc/artifacts/ask/ directory. Files are saved as markdown with the provider name, slug, and timestamp for durable record keeping.

What is the best way to send the same task to different AI providers without changing CLI commands?

Provider-agnostic advisor routing through the omc ask wrapper lets you send the same task to Claude, Codex, or Gemini via a single canonical flow. This avoids directly assembling or running provider CLIs manually.

How do I execute a security review prompt using Codex and keep the output?

Run the ask command with the codex provider and your prompt text. The wrapper routes the prompt to the local Codex CLI and automatically captures the response as a persistent markdown artifact.

Do I need to manage CLI flags manually when switching between Claude and Gemini?

No, the omc ask wrapper handles version-safe flag handling automatically. It enforces a required execution path so you do not need to directly assemble or run provider CLIs or manage their flags.

Why should I not directly assemble or run provider CLIs for advisor-style question answering?

Directly running provider CLIs bypasses the required execution path that ensures provider compatibility and artifact persistence. The omc ask wrapper enforces this path to guarantee durable record keeping for your workflows.