Give this skill to your agent

Copy the prompt and paste it into Claude Code, Codex or any agent that can read a web page. It installs the skill and tells you what it does.

Install this skill for me.

Read https://cwyz.com/skills/replica-skill-en
The raw SKILL.md is at https://cwyz.com/api/skills/replica-skill-en/raw

Save it as replica-skill/SKILL.md in my skills folder (Claude Code: ~/.claude/skills/ for every project, or .claude/skills/ for this project only).

Then tell me in two sentences what it does and when it applies. Do not run anything it describes until I say so.

Or unzip the download into ~/.claude/skills/ (every project) or .claude/skills/ in your project. A skill tells an agent what to do. Read it before you install it, especially one you did not write.

Replica Studies a Product and Builds Your Own Differentiated Version

Use when a user wants to study an existing product's features and flows and build their own version (architecture, build, testing, branding, launch). Based on the open-source replica-skill (MIT), a set of 11 Skills; only borrows features and flows, and does not copy the other party's code, logos, copy, or content. Analysis uses only public information and the user's own accounts.

Use caseHave an agent analyze a product's features and flows, then build your own differentiated version step by step.

Machine translation. Read the original

Replica Studies a Product and Builds Your Own Differentiated Version

Use this Skill when a user wants to study an existing product's features and flows and build their own version. It is built on the open-source replica-skill (MIT): a set of 11 Skills, free, with no sign-up and no API key required, linking everything from analysis to launch in one line.

Its principle is: replicate what the application does, not what it owns. Borrow only features and flows (a clean-room approach), and do not copy the other party's code, logos, copy, or content.

Triggers

Users say things like "help me analyze this app's features and flows", "build a similar product", "see what users complain about in reviews", "name my version and define its brand", "write a landing page and app store description".

Hard Rules

  1. Only analyze your own target, or a target you have explicit authorization for. If the scope of authorization is unclear, stop and ask the user before doing anything.
  2. Do not scan, attack, bypass payments or access controls, or perform denial of service against live systems. Prioritize read-only and local analysis.
  3. Tell the user before installing anything or changing configuration, and proceed only after they agree.
  4. Sort conclusions into three categories: what was directly observed (with files and locations), what was inferred, and what was not found (write "unknown", not "does not exist").
  5. Do not output or record the full contents of any keys or tokens found; only report where they are.
  6. Do not copy what the other party owns: no code, logos, copy, images, or content. For design, use open icons, open fonts, and the user's own text.
  7. Analysis uses only public pages, screenshots, app store listings, help centers, and the user's own accounts. Do not log into someone else's account, and do not touch the other party's back end or partner network.
  8. Where trademarks, patents, or terms are involved, remind the user this is not legal advice, and have a lawyer review it when necessary.

Installation

In a Claude conversation:

https://github.com/Jakeschincariol/replica-skill
install skill

Or install as a plugin in Claude Code:

/plugin marketplace add Jakeschincariol/replica-skill
/plugin install replica-skill@replica-skill

To get /replica-recon without a namespace, copy the folders instead:

git clone https://github.com/Jakeschincariol/replica-skill.git
cp -r replica-skill/replica-* ~/.claude/skills/

The bundled Python tools require Python 3.8 or later and do not require installing anything with pip. Without Claude Code, you can paste a single SKILL.md at the top of a conversation and use it as one mode, just without the Python tools.

Eleven Commands, Used in Order

Each step reads the results the previous step wrote in the project's replica/ folder.

recon → architect → design → build → backend → test → diff → entrepreneur → brand → launch → deploy
Command What it does
/replica-recon Analyze the target application: interface, flows, components, data model
/replica-architect Plan the tech stack, database structure, and interfaces
/replica-design Rebuild the design system: colors, fonts, spacing, components, using your own assets
/replica-build Build the application screen by screen, following the analysis results
/replica-backend Login, database, payments, integrations
/replica-test Click through every flow and find defects
/replica-diff Compare against the original application, giving a similarity score and missing items
/replica-entrepreneur Read real user complaints in reviews and turn them into improvement points and positioning
/replica-brand Name it and replace it with your own brand
/replica-launch Landing page, pricing, app store description
/replica-deploy Launch on your own domain

Output Format

  1. At the end of each step, state which file in replica/ was written.
  2. Scope boundaries for the analysis phase: which items are marked "out of scope" (for example, the other party's partner marketplace).
  3. Before deployment, remind the user to confirm that their brand, assets, and terms do not infringe on anyone's rights.
Done with this one? Mark it to track your path.Sign in
Save it for laterSign in
Questions about this piece?Ask AI about this

Comments (0)

No comments yet. Start the conversation.

Sign in to join the discussion.Sign in