AI Agent Hub
Back to skills
Seller Sprite ARI Amazon Review Intelligence icon

Seller Sprite ARI Amazon Review Intelligence

Data Analysis Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Install @user_507a5056/amazon-review-intelligence into your AI assistant according to https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Amazon reviews are not just buyer feedback; they are signals for product, listing, support, and competitor strategy. The hard part is turning scattered reviews into actionable judgment: which negative reviews are hurting conversion, which competitor advantages are stable, which VOC themes are worsening, and what an operations report should prioritize. Manual review reading easily misses high-vote negatives, period-over-period changes, and image-backed evidence, and it makes cost and credit checks less transparent before paid analysis.

How It Works

The skill packages the ARI CLI workflow into reviewable analysis steps. A session usually starts with check to verify the account, email verification, and usable credits. For VOC reports, run voc --site ... first to get a quote, then append --confirm after user confirmation; the command handles collection, waiting, analysis, archiving, and returns reportUrl. For day-to-day triage, use reviews --asin ... --stars negative --sort helpful for helpful negative reviews, benchmark --asin ... for category baselines, and analyze --type compare for competitor comparison. Use report --id to read archived reports; when deltaMd is returned, lead with period-over-period changes before the full text. operations supports ASIN health checks, listing audits, review-to-action workflows, and operations reports: run capabilities, profile, and quote first, then execute operations run --confirm with the same requestId after authorization. If the stream is interrupted, check operations status --request-id before deciding next steps.

Boundaries And Notes

  • Paid commands require explicit confirmation and must not be auto-confirmed.
  • Do not blindly retry 402, 429, or interrupted streams; verify whether a report was already generated before rerunning paid work.
  • Non-US sites can only use paid credits, and balance checks should rely on sufficient and usableBalance.
  • Output must separate direct data, inferred reasoning, and strategy suggestions, and preserve sample size, window, reportId, creditsUsed, and reportUrl.

Use Cases

  • Generate a quoted VOC report for an ASIN, then review themes, pain points, and improvement suggestions after confirmation.
  • Check helpful negative reviews and image-backed negatives before launch to spot product-page conversion risks.
  • Use competitor comparison and radar trends to track how this and competitor review themes change over time.
  • After a negative-review alert, use workbench and advise to locate issues and draft a response.

Best For

  • Cross-border Amazon operators who turn review issues into actionable listing fixes.
  • Sellers preparing weekly ASIN reports who prioritize by negatives, competitors, and deltas.
  • Product managers tracking conversion and returns who need VOC themes for root-cause fixes.
  • Support leads who need to triage severe negative reviews and draft response suggestions.