deck-skills

Guides Yu-Gi-Oh deck research, construction, optimization, and playtesting workflows.

1|Updated Apr 16, 2026
One-click install
npx skills add https://github.com/ZisIsNotZis/ygoskill --skill deck-skills-zisisnotzis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deck-skills
Source: https://github.com/ZisIsNotZis/ygoskill/tree/main/ygo/deck
Command: npx skills add https://github.com/ZisIsNotZis/ygoskill --skill deck-skills-zisisnotzis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires torch, lightning, numpy, pyarrow, transformers, and includes scripts (resource) and references (resource) components.

What problem does it solve? Building a competitive Yu-Gi-Oh deck requires balancing ratios, hand-trap budgets, combo engines, and matchup strategy, and doing this by intuition alone leads to bricking decks and misread metas. This Skill provides structured rules, metrics, and tactics for deck research, building, and optimization. ## Core Features & Use Cases - Deck Construction Rules and Metrics: Enforces main/extra deck size limits, card copy limits via alias and banlist checks, ydkshow ratio thresholds, and a weighted quality score formula for grading builds. - Strategy and Engine Knowledge: Documents combo engines, go-first/go-second plans, Maxx C compromise lines, interruption tactics by engine shape, and advantage/tempo theory. - Playtest and Study Workflows: Evolves deck experience through ygopromcp self-duels and Bilibili master video study, plus PyTorch-based ML utilities for deck modeling. - Use Case: Given a new deck list, validate it with ydkcheck.py, score it against the quality formula, identify its engine shape and choke points, then verify combo lines through self-duels. ## Quick Start Analyze my Yu-Gi-Oh deck list, check its construction against the rules and metrics, and suggest improvements to its ratios and hand-trap lineup.

Frequently Asked Questions about deck-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check if my Yu-Gi-Oh deck list is legal and consistent?▼

Run ydkcheck.py on the YDK file to validate deck sizes, card copy limits, and banlist compliance. The rules use card name and alias logic: aliases differing by 10 or less count as the same card, while larger differences share the 3-copy limit.

What are the recommended hand trap ratios for a competitive deck?▼

Run 9 to 12 hand traps with at least 3 types: typically Ash Blossom at 3, Maxx C at 2 to 3, Indexer at 1 to 2, and 2 to 4 others. The minimum acceptable is 6 total, but that is not recommended.

How is the deck quality score calculated?▼

The score weights core completeness at 25 percent, hand trap coverage at 20 percent, extra deck usability and first-turn impedance at 15 percent each, compromise and breakthrough at 10 percent each, and consensus match at 5 percent. Scores of 80 or above are master level.

How do I playtest a deck with ygopromcp self-duels?▼

Start srvpro, create a room with NOCHECK if the deck is not banlist-legal, and launch two ygopromcp clients with matching room and deck flags. The main agent drives the target deck while a subagent drives the opponent, logging combo lines and halt points.

What should I do when the opponent activates Maxx C?▼

Minimize special summons to reduce the opponent's draw benefit. Make a compromise end board of one negate or stall plus one standing monster, set counter traps, and use graveyard resources instead of forcing the full combo.

When should I not use the ML deck modeling utilities?▼

The llm utilities require PyTorch with CUDA, the cards.cdb SQLite database, and a YGOPro installation, so they are unsuitable without GPU resources or card data. They are reference documentation for deck modeling, not a required step for standard deck building.