agentleFS
Sign inStart free
dmminga1983dmminga1983/minga-replica/skills/replica-users/SKILL.md

replica-users

Skill1 starsChanged 18 days agoMITView on GitHub

Read what the original app's real users say in public reviews (RuStore, App Store, Google Play, Otzovik, Irecommend, Telegram, Habr, VC.ru, or G2, Capterra, Trustpilot, Reddit, Hacker News), rank what they hate, what is missing and what is unsolved, and turn it into a fix plan and a positioning angle. Every quote verbatim and linked. Use for 'what do people hate about X', 'read the reviews', 'find the gap', 'how do I position this'; RU: что бесит пользователей, разбери отзывы, найди нишу, позиционирование.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read replica/region.txt (written by /replica-recon): ru, global or both.

    line 17
  2. Distinctive:

    Never fabricate: No invented reviews, quotes, ratings, counts or sources.

    line 23
  3. Very specific:

    Verbatim and linked: Every quote copied exactly with the URL of the review or thread.

    line 25
  4. Distinctive:

    Reviewers are research, not testimonials: Never on your landing page.

    line 32

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

replica-users

A straight copy has no reason to exist. This finds the reason: what the original's users hate, in their own words, fixed in yours.

python3 reviews.py replica/reviews.csv --out replica/feedback.md

Region

Read replica/region.txt (written by /replica-recon): ru, global or both. It decides the stack, the stores, the laws and the trademark checks. python3 ../replica-architect/stack.py --region <region> prints them. No file yet: ask where the app will be sold and write it.

The rules, which are not negotiable

  • Never fabricate. No invented reviews, quotes, ratings, counts or sources. A source that cannot be reached is said to be unreachable. Fourteen reviews are fourteen.
  • Verbatim and linked. Every quote copied exactly with the URL of the review or thread. reviews.py drops rows without a URL.
  • Reading, not scraping. Review pages read the way a person reads them, rows copied into the sheet. Official feeds and APIs within their terms are fine (Apple's customer reviews RSS feed, the Hacker News Algolia API, Reddit's official API).
  • No fake reviews, ever, for your app or against theirs. Illegal in the US (the FTC rule of 2024) and in many other places.
  • Reviewers are research, not testimonials. Never on your landing page.

Step 1: collect

100 or more reviews from at least three sources, recent first, including 3 and 4 stars ("нравится, но..." is where the best fixes hide). Sources by region: ru RuStore, App Store and Google Play (Russian storefronts), Otzovik, Irecommend, Habr and VC.ru comments, Telegram chats, Yandex Maps and 2GIS for local services; global App Store, Google Play, G2, Capterra, Trustpilot, Reddit ("X alternative", "switched from X"), Hacker News, the original's feature-request board. And its changelog, so you do not "fix" what is already fixed.

Each row in replica/reviews.csv: source, url, date, rating, text.

Step 2: rank

reviews.py sorts into themes (themes.json, Russian and English, edit it for the category), weights low ratings and recent reviews, marks thin themes (under 3 reviews or one source), lists requests in the users' words and low ratings no theme matched. Read that last list by hand.

Step 3: three lists

What they hate, what is missing, what is unsolved (whole groups the app ignores: "не для команд", "useless for therapists"). Each item: the problem, how many reviews, how many sources, one or two linked quotes. Thin is labelled thin.

Step 4: fix plan and angle

The top 5 to 8 by evidence times cheapness, each with size, the skill that does it, the evidence, and a new row in features.csv with original set to no. Then three positioning lines:

For {who} who {hate this about the original, in plain words},
{your app} {does this instead}.
Evidence: {theme}, {n} reviews across {n} sources.

Recommend one. It drives /replica-brand and /replica-launch. The original's name never goes in your app name, ads or listing; a comparison page is a question for a lawyer in your country.

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