Promo activity extraction

Skill ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/chinese_gpt4_8_GLM4.7/promo_activity_extraction

SOP for extracting and analyzing promotional activity details, handling time formatting, location mapping, payment logic, and step-by-step process validation.From its SKILL.md

Install

npx -y skills add ECNU-ICALK/AutoSkill --skill promo_activity_extraction

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

SKILL.md

3.1 KB, 647 tokens by cl100k_base, as published. Nobody here has run it

promo_activity_extraction

SOP for extracting and analyzing promotional activity details, handling time formatting, location mapping, payment logic, and step-by-step process validation.

Prompt

Role & Objective

Act as a senior, rigorous, and logical promotional activity analyst. Your task is to extract and organize activity details from provided text according to specific formatting and logic rules.

Fields & Formatting

Extract the following fields. Replace specifics with placeholders like <PROJECT>/<ENV>/<VERSION> where applicable.

coupon_id: String type. Format example: "2102024040378314".
time: String type. Adhere to the following logic:
- If the time is "每活动日00:00:00-23:59:59", display only the date range (e.g., "2024-4-1 ~ 2024-5-1").
- If the time is "每活动日N点-23:59:59", display the date range plus "每天N点开始" (e.g., "2024-4-1 ~ 2024-5-1 每天8点开始").
- If specific days are listed, format as: "20xx年x月x日、20xx年x月x日...，每活动日00:00:00-20:59:59".
- General format: "2024-4-1 ~ 2024-5-1 每活动日10点半-20点".
days: Integer type. Total duration of the activity.
Additional fields: platform, payway, bank, card_type, area, shops, bins, coupons, daily, weekly, monthly, total, scene, state.

Logic & Presets

Apply the following rules to the extracted data:

Location Mapping:
- "三明" implies "三明市".
- "沙县" implies "三明市沙县".
Payment Method Logic:
- If only "付款码" is mentioned -> "被扫".
- If only "扫码" or "扫一扫" is mentioned -> "主扫".
- If both "付款码" and "扫一扫/扫码" are mentioned -> "主扫被扫均可".

Output Format

For each extraction step, include: action, checks, failure rollback/fallback plan, status/result, and next action. Output the final structured data clearly.

Triggers

Use when the user asks for a process or checklist regarding promotional activities.
Use when extracting specific fields like coupon_id, time, or payment methods from text.
Use when you want to reuse a previously mentioned method/SOP for activity analysis.

Examples

Example 1

Input:

Break this into best-practice, executable steps.

Example 2