E-commerce Product Research and Screening

Problem it solves

E-commerce product research often stalls when links, screenshots, spreadsheets, and ad-hoc notes are mixed together, while fees, platform rules, and risk signals remain scattered. This skill turns those inputs into a traceable research workflow: it confirms market, platform, price band, budget, and test window, then produces ranked candidates, profit estimates, and a minimum test plan instead of promising best-sellers.

How it works

• Mode selection by input: use discovery mode when no candidates exist; read publicly visible link data; extract screenshot fields and mark gaps; map CSV/Excel rows into batch scoring; compare text-based product notes.
• Evidence and risk checks first: review demand, competition, supply, compliance, returns, and content presentation; flag food, medical, child-safety, battery, and other high-risk categories.
• Transparent calculation: run deterministic calculations only when fee fields are complete; mark missing platform, ad, or shipping costs instead of filling zeros, or use clearly labeled optimistic/baseline/pessimistic assumptions.
• Handoff-aware execution: pause for login, captcha, authorization, payment, or irreversible actions; if Python or package files are unavailable, fall back to manual transparent scoring without pretending scripts ran.

Boundaries

It fits pre-selection research and small-batch testing planning, not bypassing logins, scraping restricted data, purchasing, contacting suppliers, or launching ads. Low evidence coverage should yield “continue research” only, and a negative pressure test should block bulk inventory. Keep sources, dates, and assumptions visible for review.