README.md

metadata

title: Pokémon Card Analyzer MCP
emoji: 🧠
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.32.1
app_file: app.py
pinned: false
license: mit
short_description: Get info from pokemon.
tags:
  - mcp-server-track
  - pokemon
  - ocr
  - api
  - agent
  - card-recognition
  - battle-simulation

🧠 Pokémon Card Analyzer MCP

Pokémon Card Analyzer MCP is a Multi-Component Program (MCP) built with Gradio that combines OCR, PokéAPI data, and AI-driven logic to analyze Pokémon trading cards and simulate battle scenarios.

With this tool, agents can recognize physical cards from images, retrieve stats, estimate levels from HP, and generate detailed info sheets. It also supports manual input — and even lets agents simulate Pokémon fights!

🎥 Demo Video

▶️ Watch the full demo

🎯 Features

• 🖼️ Image OCR Recognition – Detect Pokémon name and HP from real-world card scans or webcam images
• 🌍 Multi-language Support – Translate card names (e.g., French → English) using a custom dictionary
• 📊 Level Estimation – Estimate Pokémon level from observed HP based on official stat formulas
• 📘 PokéAPI Integration – Retrieve evolution chains, base stats, move sets, descriptions, and more
• ✍️ Manual Entry Mode – Input name + level or HP manually to simulate virtual Pokémon sheets
• ⚔️ Battle Simulation Ready – Export usable stats to plug into AI battle logic (Claude, etc.)

✨ Available Tools

• manual_mcp_hp – Estimate Pokémon level based on HP
• manual_mcp_level – View Pokémon sheet by entering level directly
• ocr_mcp – Upload or scan a card to auto-extract data and get a full PokéAPI-powered report

Each tool is accessible via an individual Gradio Interface, MCP-compatible and agent-friendly.

🛠 Tech Stack

• Gradio MCP – Modular agent interface with multi-function routing
• EasyOCR – Lightweight text recognition from images
• PokéAPI v2 – Official Pokémon REST API
• ChromaDB + OpenCLIP (optional) – Vector database for reverse image search
• Claude/OpenAI (demo) – Battle simulation logic for agent-based combat

📦 Run Locally

git clone https://github.com/your-username/pokemon-mcp
cd pokemon-mcp
pip install -r requirements.txt
python app.py