App review analyzer
Skill bg-szy/TOP-SKILLS/skills/claude-code-toolkit/app-review-analyzer
全球最大的 Claude Code 技能聚合库 · 收录 3900+ 来自 12+ 来源的技能,提供在线搜索与趋势分析看板 / The world's largest Claude Code skill aggregation hub — 3900+ skills from 12+ sources with online search and trend dashboardFrom the repository description
npx -y skills add bg-szy/TOP-SKILLS --skill app-review-analyzerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
6.4 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it
Prerequisites
- Chrome browser with Claude in Chrome extension (for reading store reviews)
- No API keys required — all analysis is done through live browser interaction
- Supports iOS App Store and Google Play Store
Mode Selection
Ask the user which mode they need:
- Competitive Analysis — Analyze a competitor's reviews to find gaps and opportunities
- Own App Management — Analyze your own app's reviews to prioritize fixes, surface feature requests, and draft professional responses
If the user provides a competitor's app, use Competitive Analysis mode. If they mention "my app" or "our app," use Own App Management mode.
Competitive Analysis Mode
Step 1: Navigate to the App Listing
Open the app's store page in Chrome:
- iOS App Store: Search on apps.apple.com or use a direct link
- Google Play: Search on play.google.com or use a direct link
Confirm the correct app with the user before proceeding.
Step 2: Collect Reviews Systematically
Read reviews in two passes:
- Most Recent — Sort by newest first. Read at least 30-50 reviews to capture current sentiment.
- Most Critical — Sort by lowest rating (1-star, then 2-star). Read at least 20-30 critical reviews to surface pain points.
For each review, note:
- Star rating
- Date posted
- Review text (key quotes)
- Whether the developer responded
Step 3: Categorize Reviews
Assign each review to one or more theme categories:
| Category | Signals |
|---|---|
| Bug Reports | Crashes, errors, data loss, freezing, sync failures |
| Missing Features | "I wish it had...", "Why can't I...", "Needs..." |
| UX Complaints | "Too complicated", "Can't find...", "Confusing", "Slow" |
| Pricing Objections | "Too expensive", "Not worth it", "Used to be free" |
| Praise | Specific features users love, "best app for...", loyalty signals |
| Support Complaints | "No response", "Unhelpful", "Can't reach anyone" |
Step 4: Produce the Analysis Report
Generate a structured report using the template in
references/analysis-report-template.md. The report must include:
- Theme Frequency Table — Count of reviews per category, sorted by frequency
- Sentiment Trend — Are recent reviews better or worse than older ones?
- Top 5 Pain Points — With direct quotes from reviews
- Top 5 Praised Features — Competitor strengths you must match or exceed
- Opportunity Summary — Gaps you could fill, weaknesses to exploit
Own App Management Mode
Step 1: Navigate to Your App's Reviews
Open your app's store page in Chrome. Confirm the correct app.
Step 2: Collect and Categorize Reviews
Follow the same collection process as Competitive Analysis (Steps 2-3 above).
Step 3: Prioritize Issues
Score each theme by Frequency x Severity:
| Severity | Definition |
|---|---|
| Critical (3) | Data loss, crashes, security issues, payment failures |
| High (2) | Core functionality broken, major UX blockers |
| Medium (1) | Nice-to-have features, minor annoyances, cosmetic issues |
Calculate priority score: count of reviews in theme x severity weight
Sort themes by priority score descending. This is the fix-first order.
Step 4: Draft Review Responses
For each negative review (1-3 stars), draft a response following the templates
in references/response-templates.md.
Key response principles:
- Respond within 24-48 hours — speed improves update likelihood
- Never be defensive or argumentative
- Personalize every response — reference specific details from their review
- Acknowledge the problem before offering solutions
- Include a direct contact method for follow-up when appropriate
- Keep responses concise (2-4 sentences for most cases)
Platform-specific notes:
- iOS App Store: Developer responses appear publicly under the review. Users receive a notification and can update their rating.
- Google Play: Developer responses also appear publicly. You can report policy-violating reviews (spam, off-topic, profanity) via the Play Console.
Step 5: Produce the Action Plan
Generate a report using references/analysis-report-template.md with an
additional action plan section:
- Fix First — Critical bugs and top pain points by priority score
- Add Next — Most-requested features with user quotes as evidence
- Communicate — Issues that need a public response or in-app messaging
- Monitor — Themes to watch in future review cycles
Review Category Quick Reference
Use this decision tree when categorizing ambiguous reviews:
Review mentions a crash/error/data loss?
→ Bug Report
Review says "I wish" or "please add" or "why can't I"?
→ Missing Feature
Review says "confusing" or "hard to use" or "can't find"?
→ UX Complaint
Review mentions price, subscription, cost, or payment?
→ Pricing Objection
Review says "no response" or "support" or "help"?
→ Support Complaint
Review is purely positive with no complaints?
→ Praise
Review has multiple themes?
→ Assign all applicable categories
Resources
references/
- response-templates.md — Detailed response templates for every review category with multiple variants each. Load when drafting review responses.
- analysis-report-template.md — Full markdown template for the analysis report output. Load when producing the final report.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most review quality skills give in ~1.3k tokens
Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07
- Ask questions one at a timein 81 of 1048, across 64 files
- Provide a recommended answer for each questionin 73 of 1048, across 50 files
- Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
- Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
- Interview the user relentlessly about the planin 38 of 1048, across 13 files
- Order findings by severityin 31 of 1048
- Resolve each branch of the decision treein 27 of 1048, across 5 files
- Run a grilling sessionin 26 of 1048, across 5 files
- Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
- Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
- Create documentation files lazilyin 24 of 1048, across 5 files
- Assign severity to every findingin 24 of 1048
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.