Amazon AI Product Selection Workflow

Problem Being Solved

Amazon product selection often gets stuck on the question of whether a specific ASIN is worth pursuing after a category looks attractive. Relying only on keyword search volume or a category Top 100 list can mistake category heat for single-product opportunity. When a reference ASIN exists, the skill needs to answer how its sales were built, whether the current price is a floor or ceiling, and whether competitors already form review and advertising barriers.

How It Works

The skill anchors on one reference ASIN and prefers SellerSprite MCP, Sorftime MCP, or user-supplied exports; it does not replace scoring data with public web results. The flow includes:
- ASIN health check: validates monthly sales, listing age, and subcategory rank to exclude weak or unrepresentative benchmarks.
- Growth-path diagnosis: uses sales, price, and rank curves, or asin_prediction, or a coupon-price fallback to identify cold start, ramp, inflection, and stable stages.
- Competitor pool and keyword reverse lookup: selects a similar-product pool, reverse-looks keywords from top competitors, and builds a demand signal closer to the actual product.
- Five-dimension scoring: scores demand density, review barrier, supply-demand ratio, differentiation room, and supply-chain difficulty, checks explainability for AI shopping assistants, then applies small-seller veto rules.
- HTML report delivery: produces an input summary, score table, action suggestions, risk notes, and next steps.

Boundary of Use

It is intended for early product screening and direction judgment, defaults to Amazon US on-site data, and does not replace final project approval involving cost, certification, supply chain, or cash-flow testing. Without a clear ASIN, the category-scorer entry should be used. If the result is not recommended or only cautious, it should not proceed directly to listing copy.