Replica skill
Eleven Claude Code skills that rebuild an app's features from scratch: recon, build, tests, review analysis, rebranding and deploy
- 6,340 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Writes the whole app's code, migrations and environment files into the project and runs builds, Playwright tests and its own scripts
- The backend and deploy skills set up auth, Stripe payments and shipping to a domain: keys and purchases stay with you, but Claude writes code and commands for production
- Reads public pages, reviews and your account in the studied app through a browser you connected
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/replicaDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Writes the whole app's code, migrations and environment files into the project and runs builds, Playwright tests and its own scripts
- The backend and deploy skills set up auth, Stripe payments and shipping to a domain: keys and purchases stay with you, but Claude writes code and commands for production
- Reads public pages, reviews and your account in the studied app through a browser you connected
This is third-party code. Review the repository files before installing.
What it does
Replica is a set of eleven skills that run as a chain and pass results to each other through a replica/ folder in your project. /replica-recon studies someone else's app from public pages, the help center, the pricing page, the store listing and your own account, and writes a map of screens, flows and components, an inferred data model and a features.csv feature matrix. /replica-architect picks the stack, the database schema and the build order, /replica-design turns screenshots into design tokens, /replica-build builds the app screen by screen and ticks off the matrix, and /replica-backend adds auth, a database, Stripe payments and email through official APIs with your keys. Then /replica-test writes a test plan and Playwright specs and logs bugs by severity, and /replica-diff compares the clone with the original and returns a parity score with a list of what is missing. /replica-entrepreneur reads the original's user reviews and turns complaints into a fix plan and a positioning angle, /replica-brand invents a name, a palette and a voice and looks for leftovers of the original, /replica-launch writes a landing page and a store listing, and /replica-deploy checks readiness and ships the app to your domain once you confirm. Six helper scripts use only the Python standard library and never touch the network: imgdiff.py compares screenshots, parity.py scores parity, reviews.py ranks complaints, contrast.py checks WCAG contrast, sweep.py finds traces of the original in code, and listing.py checks store limits.
Who it is for. For founders and independent developers who want to rebuild the features and flows of a known app under their own name, such as a niche booking service or tracker, and take it through to launch.
Good fit when
- You want your own version of an app for a narrow audience and need the original's map of screens and flows before the first line of code
- You want to learn from real reviews what the app's users complain about before deciding whether to repeat it: /replica-entrepreneur also works on its own
- A clone is built and you need a number for how much of the original's features it covers and a list of what is missing: parity.py
- You need to check that no name, domain or colors of the original remain in the code and store listing: sweep.py and listing.py
- You need a checkable path from idea to launch: stack, database schema, tests, landing page, DNS records
Not a fit when
- You need to copy an app's content, catalog, user network or licenses: the skills rebuild features and flows, and not what the original owns
- You need an exact copy of code, assets, a logo or copy: the skills write everything fresh and block deploy while traces of the original remain
- The terms of your account in the original forbid building a competitor with it: the skill then advises working from public sources only
- The product to clone is too complex for one project, such as a spreadsheet engine: /replica-recon sizes it first
Example request
Study a meeting-scheduling app I like from its public pages, build a map of its screens and flows, and estimate how long my own version would takeLimitations
- The plugin targets Claude Code, where its skills get the /replica-skill: prefix, and without it you copy the folders into ~/.claude/skills/. Without Claude Code you can paste a single SKILL.md into a chat and get the method without the Python scripts.
- The six scripts need Python 3.8 or newer, and no packages have to be installed.
- The author does not guarantee a complete copy: the result depends on the app's complexity, a booking tool takes weeks and a spreadsheet engine is out of reach. /replica-recon gives the real size estimate up front and /replica-diff gives the parity number.
- The skills read only public pages and your own account: they do not sign in to other accounts, get past paywalls or save the original's scripts and network calls.
- You handle keys, accounts, buying the domain and switching Stripe to live mode, while Claude writes variable names, DNS records and commands.
- This is not legal advice: before selling, the author advises checking trademarks, the terms of any services you used and showing the result to a lawyer.
How to disable. Disable and remove the plugin through the /plugin menu in Claude Code. If you copied the folders by hand, delete the replica-* directories from ~/.claude/skills/. The skills' results live in the replica/ folder of your project, which you remove separately.
FAQ
In what order do I run the Replica skills?
As a chain: recon, architect, design, build, backend, test, diff, entrepreneur, brand, launch, deploy. Each reads what the previous one wrote into the replica/ folder. Any of them also runs on its own, for example /replica-entrepreneur on an app you are only considering.
What does /replica-entrepreneur do with reviews?
It aims for 100 or more reviews from at least three sources: App Store, Google Play, G2, Capterra, Trustpilot, Reddit, Hacker News and the app's own public request board. You put rows with quotes and URLs into replica/reviews.csv, and reviews.py drops rows without a link and ranks themes, marking as thin those with fewer than three reviews or a single source.
How do the skills keep traces of the original out?
/replica-brand records the original's names, domains and colors in replica/brand.json, and sweep.py searches the whole codebase for them, including identifiers like OriginalAppEmbed. /replica-deploy will not ship the project until the sweep passes with exit code 0.
How is the clone's parity score calculated?
parity.py takes replica/features.csv, weights features by importance (must, should, could), counts partial as half and ignores what you left out on purpose or added beyond the original. A missing must-have means it is too early to ship.
What does /replica-deploy check before shipping?
Playwright tests, parity on must-have features, a clean sweep.py run, listing.py store limits and the npm run build production build. By hand it adds closed S1 and S2 bugs, privacy policy and terms pages, account deletion and your own favicon, titles and OG image.
Discussion
No questions or reviews yet.
Related
SkillsEleven skills for Claude Code and Codex: test audit, git worktree cleanup, plan pages, screen recording transcripts and 1Password sign-in
A Claude Code plugin: 39 commands for PRs, review, commits and specs, nine expert roles and hooks, in eight language versions
A Claude Code plugin that extracts behavioral specs, test vectors and acceptance criteria from a codebase, with a provenance trail
A Claude Code marketplace of 19 plugins and 76 skills for frontend, backend, DevOps and security, with a skill chain command
Nick Tune's plugin marketplace for Claude Code: checks before a task is handed in, code review and TDD personas
A skill for faithful rebuilds of creative sites: mirrors the page, reverse-engineers minified code line by line and verifies with automated checks
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| Replica (this entry) | no | 6,340 | |
| AI Website Cloner Template | no | 37,922 | recreates a single website from its URL as a Next.js app: screenshots, style measurement and parallel section builds |
| Web Clone | no | 1,083 | a method for reproducing websites from static pages to WebGL, relying on the real source code |
README in short
The repository's main page explains the set's purpose, gives three install routes (paste a link into the chat, install as a plugin, copy the folders), a table of eleven commands and a walkthrough of the chain on a meeting-booking app. It then describes the six scripts with run examples, a section on boundaries (clean room, public sources only, mandatory rebranding), a file list and the MIT license. The repository holds the folders of all skills with templates and examples and a tests directory with tests for each script.
README badge
Original SKILL.md
--- name: replica-recon description: >- Reverse-engineers any app into a recon map: screen inventory, user flows, component list, inferred data model and a feature matrix, from public pages, screenshots, app store listings, help docs and the user's own account. The first step of the Replica pack. Use when the user says "clone this app", "reverse engineer X", "how does X work", "map out X", "what screens does X have", "I want to build my own version of X", "copy this app", or pastes an app's URL or App Store link and wants to rebuild it. --- # replica-recon Everything else in the pack builds from what this skill writes. A bad recon map means a bad clone, so take the time here. Output goes in the user's project: ``` replica/recon.md the recon map (template: recon-map.md in this folder) replica/features.csv the feature matrix (template: features.csv in this folder) replica/screens/ reference screenshots of the original. Never shipped. ``` ## The rules, before anything else This skill rebuilds **functionality and UX patterns**, clean-room style. It studies what the app does and how a user moves through it. It does not take anything the app owns. - **Public sources and the user's own account only.** Never log into an account that is not the user's, never ask for a password, never get past a paywall or a login by any trick. - **Reading, not scraping.** No crawlers, no bulk downloads, no loops. If a browser tool is connected to the user's own browser, read pages at human speed with the user present. - **No source code, no private APIs.** Do not read or save the app's JavaScript bundles, decompile its binary, or log its network calls to copy endpoints. Public API docs are fine to read. - **Check the terms.** Some products' terms forbid using an account to build a competing product. If the user's account is under terms like that, say so and work from public sources only. - **Screenshots are reference.** They live in `replica/screens/`, are used to compare layouts, and never go into the clone.