ask

Queries local Claude or Gemini advisor CLIs and saves reusable markdown artifacts.

Updated Feb 1, 2024
One-click install
npx skills add https://github.com/ogiboy/portfolio-app --skill ask-ogiboy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ask
Source: https://github.com/ogiboy/portfolio-app/tree/main/.codex/skills/ask
Command: npx skills add https://github.com/ogiboy/portfolio-app --skill ask-ogiboy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Getting a second opinion from an external AI advisor usually means copy-pasting context between tools and losing the results. This Skill routes focused questions to a locally installed Claude or Gemini CLI and preserves every exchange as a reusable markdown artifact. ## Core Features & Use Cases - Backend Selection: Choose between the Claude or Gemini local CLI based on the request, replacing the older separate ask-claude and ask-gemini skills. - Local-First Execution: Runs the installed CLI binary directly and never silently falls back to an MCP or remote provider when the binary is missing. - Artifact Capture: Saves each session to .omx/artifacts/ask-<backend>-<slug>-<timestamp>.md with the original task, final prompt, raw CLI output, summary, and action items. - Use Case: Ask Claude to review a tricky function for edge cases, then keep the full transcript and next steps in the artifacts folder for later reference. ## Quick Start Ask the local Claude CLI to review my authentication module for security issues and save the response 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 ask Claude or Gemini a question from the command line?

Run `omx ask claude "your question"` or `omx ask gemini "your question"`, or invoke the CLIs directly with `claude -p` or `gemini -p`. The skill selects the backend based on your request and executes the local binary.

What is the difference between the Claude and Gemini backends?

Both backends run locally installed advisor CLIs for questions, reviews, and brainstorming. Use claude when you want Anthropic's model and gemini for Google's model; if unspecified, the skill picks the installed backend matching your request.

What happens if the local Claude or Gemini CLI is not installed?

The skill explains that a local CLI is required and does not silently switch to an MCP or remote provider. You must install the claude or gemini binary before the skill can execute.

Where are the advisor responses saved?

Each execution saves a markdown artifact to `.omx/artifacts/ask-<backend>-<slug>-<timestamp>.md`. It contains the original task, the final prompt, raw CLI output, a concise summary, and action items.

Can this skill use a remote API instead of the local CLI?

No. Local execution is the default and required path. The skill explicitly avoids switching to MCP or remote providers when the local binary is missing, keeping execution on your machine.