Jakeschincariol/replica-skill

Eleven free Claude skills that clone any app: reverse-engineer it, rebuild it, test it for bugs, then fix what its users hate. Free, MIT.

What it solves

Replica is a set of eleven Claude skills designed to automate the process of reverse-engineering and rebuilding an existing application's functionality and user experience patterns. It allows users to create a "clean-room" clone of an app—rebuilding the features, flows, and design systems from scratch without copying source code, proprietary assets, or private APIs.

How it works

The project consists of a series of specialized skills (modes) that are executed in a sequence to move from analysis to deployment. Each skill reads the output of the previous one, storing data in a replica/ folder within the project.

  1. Analysis & Planning: /replica-recon reverse-engineers the app's screens and flows from public pages and screenshots; /replica-architect plans the tech stack and database schema.
  2. Design & Build: /replica-design creates a design system based on tokens; /replica-build implements the screens and features.
  3. ** Backend & Testing**: /replica-backend handles auth and payments; /replica-test performs bug testing via Playwright specs.
  4. Optimization & Branding: /replica-diff scores parity with the original; /replica-entrepreneur analyzes real user reviews to identify pain points and improvements; /replica-brand handles naming and rebranding.
  5. Launch: /replica-launch and /replica-deploy handle the landing page, store listings, and production deployment.

The system includes six Python-based tools for layout diffing, parity scoring, review analysis, WCAG contrast checking, codebase sweeping for original assets, and store listing validation.

Who it’s for

Entrepreneurs and developers who want to build a competing application by studying existing successful patterns and improving upon them based on real user feedback.

Highlights

  • Clean-room implementation: Strictly rebuilds functionality and UX patterns, not source code or content.
  • Agentic workflow: A structured 11-step pipeline from reconnaissance to deployment.
  • User-centric improvement: Includes a specific skill for analyzing real-world reviews to find a competitive angle.
  • Zero-dependency Python tools: Includes specialized tools for image diffing and parity scoring using only the standard library.

Related

  • Project
    VoltAgent/awesome-codex-subagents

    This is a repository of ready-made AI subagent definitions for OpenAI's Codex, covering a wide range of development tasks with smart model selection and sandbox controls.

  • Project
    twostraws/SwiftUI-Agent-Skill

    An agent skill for AI coding assistants that provides specialized knowledge to help them write modern, performant, and accessible SwiftUI code while avoiding common LLM mistakes.

  • Project
    kucherenko/jscpd

    A high-performance Rust-based engine that detects duplicate code, dead code, and complexity hotspots across 224+ languages, featuring AI-ready integration via an MCP server.

  • Project
    getsentry/MobileBuildMCP

    MobileBuildMCP is an open‑source tool that runs an MCP server and a CLI to let AI agents (or humans) invoke Xcode build and test commands for iOS/macOS projects. Install via Homebrew or npm, start the server with `mobilebuildmcp mcp`, and use subcommands like `simulator build` to compile or test apps. Optional “skills” JSON files teach LLM assistants how to talk to the server. MIT‑licensed, macOS‑only, with Sentry telemetry.