mcpField
T
theory-analysis-purchase-motivation

theory-analysis-purchase-motivation

github.com/timlai666Marketing & growth045 installsUpdated 1d ago
Repository

About theory-analysis-purchase-motivation

A marketing skill that analyzes evidence from interviews, support tickets, social posts, and other sources through purchase motivation theory. It labels functional, security, and relational motives, with evidence checks and optional handoff to `review-mining-stp`.

Key features

  1. 01Codes functional, security, and relational motives
  2. 02Checks evidence sufficiency before analysis
  3. 03Requires auditable quotes for conclusions
  4. 04Outputs JSON and Markdown
  5. 05Supports optional `review-mining-stp` handoff

Use cases

  1. 01Identify purchase drivers in interview summaries
  2. 02Compare motives across support and social data
  3. 03Prepare theory annotations for `review-mining-stp`

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/timlai666/skills --skill theory-analysis-purchase-motivation

After installing, type/theory-analysis-purchase-motivationin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

agents/
openai.yaml
references/
00-theory-foundation.md
01-input-and-gate.md
02-dimension-coding-rules.md
03-output-contract-json-markdown.md
04-worked-example-turbocharger.md

Details

From the project README

Theory Analysis Purchase Motivation

Overview

用購買動機理論(Purchase Motivation Theory)分析跨來源證據中的決策驅動因素。 來源可為訪談摘要、客服工單、社群貼文、研究筆記、觀察紀錄等。 本 skill 含完整理論章節,可在分析前先校準動機理論與判準邊界。 此 skill 可獨立使用,也可選擇將輸出銜接 review-mining-stp。

Input Contract

至少提供:

  • analysis_goal: string
  • evidence_items: array
    • item_id: string
    • content: string
    • content_type: string

建議可選:

  • source_type: string
  • source_ref: string
  • context_tags: string[]

Data Sufficiency Gate

若缺必要欄位或沒有可用證據內容,輸出 MissingDataOutput,不做硬判讀。

{
  "missing_fields": [],
  "why_needed": {},
  "questions_to_user": [],
  "next_step_rule": "Provide analysis_goal and at least one evidence item with item_id, content, and content_type, then rerun coding."
}

Workflow

  1. 先讀 references/00-theory-foundation.md,校準理論概念、核心動機與研究應用語境。
  2. 讀 references/01-input-and-gate.md,完成輸入檢核。
  3. 讀 references/02-dimension-coding-rules.md,依三種動機做證據標註:
    • functional
    • security
    • relational
  4. 讀 references/03-output-contract-json-markdown.md,輸出 JSON + Markdown。
  5. 需要範例時讀 references/04-worked-example-turbocharger.md。

STP Integration

  • theory_annotations.family 固定為 purchase_motivation
  • theory_annotations.subtheory 只可使用:
    • functional
    • security
    • relational

輸出欄位可直接轉進 review-mining-stp 的理論標註層。 review-mining-stp 屬於 optional downstream,不是主流程前提。

Quality Rules

  • 每個動機結論必須至少有一則可稽核引文,否則標記 insufficient。
  • 不能把運送、包裝、售後都直接歸到 functional,需依定義判斷。
  • taxonomy 必須完全符合既有規則,不得自創 subtheory。

Suggested Prompts

  • Use $theory-analysis-purchase-motivation to label purchase drivers from mixed evidence items and output JSON + Markdown in Traditional Chinese.
  • 請用 $theory-analysis-purchase-motivation 分析多來源證據中的購買動機,並產出可選串接 review-mining-stp 的 theory_annotations。

Comments

Markdown supported