agentleFS
Sign inStart free
AppmanAI-comAppmanAI-com/AppmanAi-MCP/skills/review-mining/SKILL.md

review-mining

Skill3 starsChanged 50 days agoMITView on GitHub

Pull an app's reviews and turn them into grouped, sized themes — the recurring complaints and requests behind a rating, with the rating history that shows when they started. Use when asked what users complain about, why a rating fell, or for feature requests from reviews. Requires the AppmanAi MCP connector.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. shawnpang/startup-founder-skills★ 331same name
    2. Maudeunfledged834/startup-founder-skills★ 8same name
    3. BenjaminG/ai-skills★ 4same name
    4. fjilvvi/growthlens★ 2same name
    5. TomSzenessy/AI_TEMPLATE★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep themes concrete. "Bad UX" is not a theme.

    line 30

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review mining

Turn a wall of reviews into a short list of things worth fixing, sized by how often they come up.

Inputs

  • app — package id or name
  • store — ios or android
  • country — two-letter code, default US

Cost

Roughly 15–25 credits.

Steps

1 · Resolve the app. app_search (2) if you were given a name.

2 · Read the shape of the rating first. app_metrics_ratings_worldwide (6) for the star histogram. A 4.2 built from 5s and 1s is a different problem from a flat 4.2 — the first is a broken segment, the second is mediocrity. Say which one this is.

3 · Pull reviews. app_reviews (4). Take the largest sample the tool will return.

4 · Group into themes. Cluster by what the user is actually complaining about, not by wording. For each theme record: how many reviews, the star ratings they carry, the date range, and one representative quote.

Keep themes concrete. "Bad UX" is not a theme; "can't find the export button" is.

5 · Date the themes. app_metrics_ratings_history (5). Line the theme date ranges up against the rating curve. A complaint that starts the week the rating fell is your headline; an evergreen one that has run for a year is a different kind of problem.

6 · Report. Themes ordered by count, each with: the theme, the count and share, the typical star rating, when it started, and one quote. Then a short paragraph separating what is dragging the rating right now from what is chronic.

Rules

  • Size every theme. "Users complain about sync" is worthless without "in 14 of 60 reviews, all 1–2 star, starting in March".
  • One sample, one country. Reviews are per-country; do not blend markets in one theme list without saying so.
  • Quote, don't paraphrase. One real sentence per theme, verbatim.
  • Separate volume from severity. Twenty complaints about a cosmetic bug and three about data loss are not ranked by count alone — flag the severe ones explicitly.
  • Don't invent causes. Reviews tell you what users noticed, not why it happened.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free