Claude Code Skill · E-Commerce Research

AI E-Commerce
Signal Engine.

Turn thousands of customer voices into a scored e-commerce decision report in about 30 minutes. Built for operators who need sourcing, positioning and product judgement before they spend money.

Amazon reviewsReddit voiceGoogle TrendsFBA cost ceiling
$59.99one-time · instant Gumroad delivery

Launch product for solo Shopify, Amazon FBA and TikTok Shop operators.

signal-engine // scored-report
AI E-Commerce Signal Engine cover showing a scored signal dashboard

See it work in 50 seconds

How it works

Brief. Process. Decide.

01

Brief the niche

Type a product niche, known competitor or buyer problem. Optionally send a product photo through the Telegram bridge.

02

Process the signals

The engine gathers customer voice, demand movement, competitive gaps and cost constraints into one structured run.

03

Read the verdict

You receive a report, dashboard and mind-map with frequency-scored wedges, two product concepts and a plain decision.

What's inside

Four signals. One verdict.

The pack turns fragmented research sources into a consistent operator workflow: customer voice, market demand, economics and product concepts in one run.

01

Amazon review voice

Mines 3-star-weighted reviews across benchmark products so the report sees real buyer friction, not only happy testimonials.

Customer voice
02

Reddit language

Pulls unfiltered buyer language from relevant communities to reveal pains, objections and positioning words customers already use.

Market language
03

Google Trends demand

Adds a demand trajectory chart so you can separate rising categories from noisy niches that only look exciting in isolation.

Demand signal
04

Backward FBA economics

Turns a target retail price into a maximum factory price, so sourcing decisions are grounded before you brief a supplier.

Unit economics
What you get

A research system, ready to run.

  • Portable Claude Code and Codex skill for running the Signal Engine.

  • HTML report and dashboard templates for grounded customer-voice output.

  • Amazon, Reddit, YouTube and Google Trends source workflow notes.

  • Backward FBA calculator logic for retail-to-FOB sourcing checks.

  • Sample reports and setup materials packaged for Gumroad delivery.

Built for operators who need a sharper decision before sourcing.

Use it before buying samples, briefing Catherine, writing listings or committing to paid traffic. The aim is a clean go, validate or reject decision.

FBA sourcing check

Know the factory-price ceiling before you brief suppliers.

The backward FBA calculation links retail price, marketplace costs and sourcing reality. That turns the report from interesting research into a practical sourcing filter.

Report outputs

Voice wedges

Recurring customer complaints converted into scored differentiation angles.

Demand chart

Google Trends movement summarised in the report, not left as another open tab.

FBA ceiling

A practical maximum factory-cost target based on the price you need to sell at.

Concept proposals

Two product directions with sourcing, positioning and copy implications.

FAQ

Before you buy

Is this a list of product ideas?

No. It is a research engine for evaluating a niche you already care about. The output is a decision report, not a generic trend list.

Does it replace judgement?

No. It compresses noisy customer and demand signals into a structured verdict so you can make a better operator decision faster.

Do I need to code?

You do not need to write code, but you should be comfortable following setup steps in Claude Code or Codex and adding your own API keys.

What happens after purchase?

Gumroad gives instant access to the pack. Follow the setup guide, install the skill, add the required keys and run your first niche.

Make the product decision before the product spend.

A portable agentic research engine for solo e-commerce operators. Customer voice, demand, economics and concepts in one scored report.

$59.99one-time · instant access

Secured by Gumroad · Digital product delivered instantly