Intelligent E-commerce Product Selection Assistant

Problem Being Addressed

Sellers often face product-selection decisions that look precise but cannot be verified: inflated monthly sales, star ratings without scoring dimensions, and generic advice that ignores platform rules, supply-chain costs, compliance, and after-sales risk. This skill is intended to turn vague category research into actionable product decisions, such as shortlisted sub-niches, scoring rationale, price bands, validation tasks, and test-selling recommendations.

How It Works

The skill first confirms key inputs such as platform, category, price band, seller stage, sourcing capability, minimum order quantity, logistics capability, and analysis depth. If information is missing, it asks before assuming. It then routes requests into M1 category selection, M2 competitor analysis, M3 profit calculation, M4 keyword analysis, M5 hit evaluation, or M6 multi-product comparison. Full selections cover market opportunity, competition, product fit, and supply-chain feasibility, using a weighted scoring model. Profit calculation requires inputs such as sale price, product cost, platform, and shipping mode, then outputs gross margin, per-unit profit, break-even ROI, and sensitivity analysis.

Boundaries And Notes

It should not be treated as a live market-data API. When no real-time source is available, it should mark figures as estimated, inferred, or to be verified instead of inventing exact sales, search indices, or growth rates. If a competitor link cannot be accessed, analysis must rely on user-supplied information or comparable products. Compliance risks such as food, cosmetics, batteries, children’s products, and IP issues should be called out explicitly. Final recommendations should include validation actions such as sourcing quotes, checking the top 20 search results, reviewing main-image homogeneity, and running a small test order.