FlowCay logo
FlowCay
Simple Review Analyzer Skill

Simple Review Analyzer Skill

AI-powered e-commerce review analyzer with 22-dimension tags, user personas, VOC insights and visual dashboards.

Install command

npx skills add https://github.com/liangdabiao/simple-review-analyzer
SkillOpen SourceAll Platforms
Added on Popularity 3

Details

Overview

Simple Review Analyzer is an open-source (MIT) AI-powered e-commerce product review analysis skill for Claude Code CLI and OpenClaw. It automatically extracts 22 dimensions of smart tags across 8 categories — covering user personas, usage scenarios, feature satisfaction, quality, service, experience, market and sentiment — and produces CSV label data, a Markdown insights report and an interactive HTML dashboard.

Key Features

  • 22-dimension tagging: auto-classifies review content across 8 categories — audience (gender, age, occupation, buying role), scenario, feature satisfaction, quality (material, workmanship, durability), service (shipping, packaging, support, returns, warranty), experience (comfort, usability, design, price perception), market (competitor comparison, repurchase intent) and sentiment.
  • Three-part output:
  1. CSV label data — raw reviews + 22-dimension AI tags
  2. Markdown insights report — strategic opportunities, pain points, and prioritized improvement recommendations
  3. HTML visual dashboard — 6 interactive Chart.js charts, color-blind-friendly palette, responsive and print-friendly
  • Rich report structure: 10 sections including persona & scenario analysis, satisfaction attribution, negative root-cause analysis, VOC deep-dives on representative users, and strategic data output (moat, weaknesses, execution matrix).
  • Information density scoring: each review gets an info_score (1–20) based on length, valid tags, competitor mentions, repurchase intent and scenario descriptions.
  • Natural language use: invoke directly in Claude Code/OpenClaw, e.g. "analyze this product's reviews: reviews.csv".
  • Fuzzy column matching: accepts Chinese or English headers for content, rating, date, title and username.

Requirements

  • Python 3.7+
  • Claude Code CLI or OpenClaw for running the analysis skill

Usage Examples

  • Specify a file: "please analyze this product's reviews: reviews.csv"
  • Competitive analysis: "do a deep analysis of competitor reviews #reviews.csv"
  • Limit quantity: "analyze the most recent 100 review rows #reviews.csv"

Output

Analysis generates a dated output folder (e.g. output/{Product}_{YYYYMMDD}/) containing:

  • reviews_labeled.csv — original reviews + 22-dimension labels
  • 分析洞察报告.md — deep insights report with suggestions
  • 可视化洞察报告.html — high-quality visual dashboard

CLI Tools

  • csv_processor.py: export batch JSON (--export-batches -b 30) or merge tags to CSV (--merge-tags --stats)
  • transform_logic.py: JSON↔CSV conversion in both directions

Use Cases

  • Product optimization: surface pain points and improve design/features
  • Competitive analysis: learn competitor strengths and find differentiation
  • Market research: batch-analyze reviews for demand and category trends
  • User insight: build precise customer personas and refine marketing strategy

Best Practices

  • ~100 reviews recommended for a good speed/quality balance
  • Maximum 30 reviews per batch for concurrency limits
  • Auto-detects UTF-8/GBK/GB2312 encodings

License

MIT. Free and open source.

How to Use

  1. Use a Skill-compatible agent such as Claude Code, Cursor, or Codex.
  2. Install from the terminal (Node.js required):
npx skills add https://github.com/liangdabiao/simple-review-analyzer

Or paste the install prompt from this tool page into your agent and ask it to install for you. 3. Restart / reload the agent so the new Skill is loaded. 4. Invoke the Skill with its slash commands or by asking the agent to run the workflow described in Overview / Key Features. 5. Review outputs before publishing — treat model-generated copy, images, and data as drafts.