A real time battle decision tool for Pokemon Champions (2026) built in Python with a C++ damage calculation engine.
Pre-Match (1 minute window):
- Input opponent's 6 Pokemon
- Loads your saved box automatically
- Scores each of your pokemon against opponent team
- Recommends best 4 to bring with type coverage reasoning
In-Battle (40 seconds window):
- Input your active Pokemon + current HP
- Input opponent's active Pokemon + current HP
- Enter your 4 moves
- Recommends best move with damage estimate, effectiveness, speed check
Opponent's team: Charizard [Fire/Flying] Venusaur [Grass/Poison] Blaziken [Fire/Fighting] Abomasnow [Grass/Ice] Pikachu [Electric] Lucario [Fighting/Steel]
Your team score: Infernape [Fire/Fighting ] score:19.75 Ceruledge [Fire/Ghost ] score:18.25 Froslass [Ice/Ghost ] score:14.0 Pelipper [Water/Flying ] score:13.0 Scizor [Bug/Steel ] score:8.75 Raichu [Electric ] score:8.75
BRING THESE 4 POKEMON <<< Infernape (score: 19.75) Ceruledge (score: 18.25) Froslass (score: 14.0) Pelipper (score: 13.0)
=== BATTLE ADVISOR === Your : Infernape HP: 76/76 Opponent: Abomasnow HP: 90/90 ⚡ You go FIRST (spd 108 > 60)
Move analysis: flare-blitz dmg:451 effectiveness: 4.0x (double super effective) heat-wave dmg:319 effectiveness: 4.0x (double super effective) flamethrower dmg:303 effectiveness: 4.0x (double super effective) solar-beam dmg:33 effectiveness: 0.5x (not very effective)
USE: flare-blitz (double super effective, ~ 451 damage) <<<
- Python 3.11+
- g++ compiler (for C++ damage engine)
- Clone the repository
git clone https://github.com/amlan-sinha07/pokemon_advisor.git
cd pokemon_advisor- Install Python dependencies
pip install requests- Compile C++ damage calculator
g++ engine/damage_calc.cpp -o engine/damage_calc- Data is already cached — 1025 Pokemon ready to use.
Run as normal user (no admin required):
python main.py- Pre-match team selector = use before match starts
- In-battle move advisor = use during battle
- Manage my box = save your pokemon's once
- Exit
Go to option 3 first — add your 6 Pokemon to your box. After that, option 1 loads your team automatically.
User input
↓
Python (data layer)
1025 Pokemon cached from PokeAPI Move data fetched on demand + cached locally Fuzzy name matching (partial names work)
↓
Python (calculation layer) Type effectiveness (18×18 chart) Team scoring algorithm
↓
C++ (damage engine) Standard Pokemon damage formula Called via subprocess for speed
↓
Recommendation output
- Python 3.11 — core logic, CLI, data pipeline
- C++ — damage calculation engine
- PokeAPI — Pokemon and move data source
- JSON — local data caching
- difflib — fuzzy name matching
- subprocess — Python-C++ bridge
- PokeAPI — base Pokemon stats and moves
- Type effectiveness chart — hardcoded from official Pokemon Champions data
pokemon_advisor/
├── data/ ← cached JSON data
├── engine/ ← type calc, damage calc (Python + C++)
├── models/ ← Pokemon, Move classes
├── scraper/ ← PokeAPI data fetcher
├── box_manager.py ← personal box save/load
├── data_loader.py ← loads data into objects
├── main.py ← CLI entry point
└── my_box.json ← your saved Pokemon box
Amlan Sinha
GitHub: amlan-sinha07
Built in 32 days as part of a self-directed Python + C++ learning journey.