build-tcg-deck
About
This Claude Skill helps developers build and optimize trading card game decks for competitive or casual play across games like Magic: The Gathering and Pokémon TCG. It handles archetype selection, mana curve analysis, win condition identification, and sideboard construction. Use it when creating new decks, adapting to meta-game changes, or evaluating new card sets for tournament readiness.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/build-tcg-deckCopy and paste this command in Claude Code to install this skill
Documentation
Build TCG Deck
Construct TCG deck archetype → final optimization. Works across Pokemon TCG, MTG, FaB, other major TCGs.
Use When
- New deck for tournament format or casual
- Adapt existing deck to changed meta
- Eval whether new card/set warrants change
- Teach deck construction principles
- Convert concept → tournament-ready list
In
- Required: Game (Pokemon TCG, MTG, FaB, etc.)
- Required: Format (Standard, Expanded, Modern, Legacy, Blitz, etc.)
- Required: Goal (competitive, casual, budget)
- Optional: Preferred archetype (aggro, control, combo, midrange)
- Optional: Budget constraints
- Optional: Current meta (top decks, expected field)
Do
Step 1: Define Archetype
Choose strategic identity.
- ID available archetypes in format:
- Aggro: Early pressure + efficient attackers
- Control: Answer threats, win late via card advantage
- Combo: Assemble card combos → powerful synergy / instant wins
- Midrange: Flexible, shifts aggro ↔ control
- Tempo: Resource advantage via efficient timing + disruption
- Select based on:
- Playstyle
- Meta positioning (what beats top?)
- Budget (combo needs specific expensive)
- Format legality (bans, rotation)
- ID 1-2 primary win conditions:
- How does deck actually win?
- Ideal game state to reach?
- State archetype + win condition clearly
→ Clear archetype + win conditions. Specific enough to guide selection, flexible to adapt.
If err: No archetype feels right → start w/ strongest individual cards, let archetype emerge from pool. Sometimes best deck built around a card, not concept.
Step 2: Build Core
Select cards defining strategy.
- Core engine (12-20 cards depending on game):
- Directly enable win condition
- Max legal copies
- Non-negotiable — deck fails w/o
- Support (8-15):
- Find/protect core
- Draw/search for consistency
- Protection (counters, shields, removal)
- Interaction (8-12):
- Removal for opponent threats
- Disruption for opponent strategy
- Defensive opts appropriate to format
- Resource base (game-specific):
- MTG: Lands (24-26 for 60-card, 16-17 for 40-card)
- Pokemon: Energy (8-12 basic + special)
- FaB: Pitch value distribution (balance red/yellow/blue)
→ Complete list at/near min deck size. Every card has role (core, support, interaction, resource).
If err: Exceeds format size → cut weakest support first. Core needs too many (>25) → strategy too fragile, simplify win condition.
Step 3: Analyze Curve
Verify resource distribution supports strategy.
- Plot mana/energy/cost curve:
- Count cards at each cost (0, 1, 2, 3, 4, 5+)
- Match archetype:
- Aggro: peaks 1-2, drops after 3
- Midrange: peaks 2-3, moderate at 4-5
- Control: flatter, more high-cost finishers
- Combo: concentrated at combo-piece costs
- Check color/type distribution (MTG: color balance; Pokemon: energy coverage):
- Resource base can reliably cast on curve?
- Color-intensive cards need dedicated support?
- Verify card type balance:
- Enough creatures/attackers for pressure
- Enough spells/trainers for interaction + consistency
- No critical category missing
- Adjust if curve doesn't support
→ Smooth curve → deck executes strategy on time. Aggro fast, control survives early, combo assembles on schedule.
If err: Lumpy (too many expensive, not enough early) → swap expensive support for cheaper. Curve > any individual card.
Step 4: Meta Positioning
Eval vs expected field.
- ID top 5 decks in current meta (tournament results, tier lists)
- Each top deck:
- Favorable: Strategy counters theirs (+1)
- Even: No structural advantage (0)
- Unfavorable: Theirs counters yours (-1)
- Calc expected win rate vs field:
- Weight by opponent meta share
- 60%+ vs top 5 = well-positioned
- Poor positioning → consider:
- Switch interaction to target worst matchups
- Sideboard (if format allows) for unfavorable
- Whether diff archetype better positioned
→ Clear picture of where deck sits. Favorable + unfavorable matchups ID'd w/ specific reasons.
If err: Meta data unavailable → focus on versatility, interact w/ multiple strategies vs optimizing for one matchup.
Step 5: Sideboard
Construct sideboard/side deck for format adaptation (if applicable).
- Each unfavorable matchup (Step 4):
- 2-4 cards significantly improve
- High-impact, not marginal
- Each sideboard card, know:
- What matchup(s) it comes in against
- What it replaces from main
- Whether bringing it changes curve significantly
- Verify sideboard ≤ format limits (MTG: 15, FaB: varies)
- No sideboard card only relevant vs one fringe deck
- Each slot covers ≥2 matchups if possible
→ Focused sideboard meaningfully improves worst matchups w/o diluting main.
If err: Sideboard can't fix worst matchups → deck poorly positioned in meta. Core strategy may need adjust, not sideboard patches.
Check
- Archetype + win conditions clearly defined
- Format legality met (bans, rotation, card count)
- Every card has defined role (core, support, interaction, resource)
- Curve supports strategy speed
- Resource base reliably casts on curve
- Meta matchups evaluated w/ specific reasoning
- Sideboard targets worst matchups w/ clear swap plans
- Budget satisfied (if applicable)
Traps
- Too many win conditions: 3 ways to win → none done well. Focus 1-2
- Curve blindness: Powerful expensive cards w/o checking if deck casts on time
- Ignore meta: Building in vacuum. Best in theory loses to most common in practice
- Emotional inclusion: Pet card not serving strategy. Every slot earns place
- Sideboard afterthought: Last w/ leftover. Sideboard = part of deck, not appendix
- Over-teching: Narrow answers to specific decks vs proactive strategy
→
grade-tcg-card— card condition assessment for tournament legality + collection valuemanage-tcg-collection— inventory mgmt for tracking which cards available
GitHub Repository
Related Skills
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
polymarket
MetaThis skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.
creating-opencode-plugins
MetaThis skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
