Mk 358

---
name: mk-358
description: "AI Search Test Question Bank. Use when work requires definition, comparison, context, and decision questions; process the specified inputs using this method and verify the deliverable"
---

# MK-358 | AI Search Test Question Bank

For ordinary chat, upload `packs/mk-358.md`, which includes the shared rules, this method, and required references

## Purpose

This skill delivers: **Definition, comparison, context, and decision questions**

One answer does not establish a stable ranking; do not simulate current results when actual answers have not been obtained

## Required inputs

Provide: Approved brand/product/service facts, customer decisions and question sources, test market and purpose; existing questions help deduplication, while AI answers and citations are recorded after testing

## Method steps

1. Build an AI search question bank covering at least definitions, comparisons, methods, contexts, recommendations, and decisions
2. Questions should simulate real user needs, rather than testing only the brand name
3. The course assignment starts with a 30-question bank, then selects the most important 10 for live testing

## Calculation and interpretation

Retain questions, dates, platforms, and actual answers; one result does not represent answers shown to all users

## Required output fields

| Order | Field | How to complete |
|---:|---|---|
| 1 | Branded test question | Use the supplied source, calculation, or explicit unknown |
| 2 | Customer decision and source | Use the supplied source, calculation, or explicit unknown |
| 3 | Question type and observation target | Use the supplied source, calculation, or explicit unknown |
| 4 | Priority | Use the supplied source, calculation, or explicit unknown |
| 5 | Test platform and conditions | Use the supplied source, calculation, or explicit unknown |
| 6 | Not-yet-tested status | Use the supplied source, calculation, or explicit unknown |

## Verification cases

**MK-358-A1 | Brand facts/customer questions exist, no AI test yet**

Expected: Deliver branded questions/test conditions, leaving responses untested

## Method sources

Original file: SEO 106｜Check 成效診斷：從 AI 回答結果看出內容與品牌缺口

Source locator: questions

Source basis: extracted from course materials

Original URL: https://docs.google.com/presentation/d/1yJsbQNDgX4lUtExwfxXgaPvHjEwbjywvCKllchkIoOU/edit

English translation of source excerpt retained in the master sheet:

> • In what types of questions does it appear?