Game Concept — Ideation & Vision
Generate and refine game concepts from a spark of an idea into a structured pitch ready for full design.
Inputs
Gather whatever the user has — even a single sentence is enough to start:
- Inspiration: A theme, mechanic, feeling, or reference game
- Platform targets: PC, console, mobile, VR, web
- Scope: Solo jam game, indie release, or large-scale production
- Team size & skills: What's realistically buildable
If the user has nothing yet, run a guided brainstorm using the prompts below.
Step 1: Idea Generation
If the user needs help finding an idea, work through these lenses:
Genre × Theme Matrix
| Sci-Fi | Fantasy | Horror | Slice-of-Life | Historical | |
|---|---|---|---|---|---|
| Platformer | Gravity shifts | Enchanted forests | Haunted mansion | Rooftop courier | Ancient ruins |
| Roguelike | Space station | Dungeon crawl | Eldritch depths | Office survival | Time periods |
| Strategy | Colony sim | Kingdom builder | Cult management | City planner | War campaigns |
| Puzzle | Hacking | Alchemy | Escape room | Cooking | Archaeology |
| RPG | Cyberpunk | Classic quest | Survival horror | Life sim | Dynasty |
Brainstorm Prompts
Ask the user 3-5 of these:
- What game do you wish existed but doesn't?
- What's a mechanic you love? What if it was the entire game?
- What non-game experience would make a great game? (cooking, hiking, archaeology)
- Take two genres that don't normally combine — what happens?
- What emotion do you want the player to feel most?
- What's the "one more turn / one more run" hook?
- If you had to describe the game in a single verb, what is it?
Step 2: Concept Pillars
Define 3-4 design pillars — the non-negotiable truths of the game:
Example — "Void Courier" (Sci-Fi Delivery Roguelike):
┌─────────────────────────────────────────────────┐
│ Pillar 1: MOMENTUM │
│ → The player should always be moving forward │
│ │
│ Pillar 2: RISK vs REWARD │
│ → Every delivery route is a gamble │
│ │
│ Pillar 3: EXPRESSIVE LOADOUTS │
│ → Ship customization changes how you play │
│ │
│ Pillar 4: EMERGENT STORIES │
│ → No two runs tell the same story │
└─────────────────────────────────────────────────┘
Step 3: Elevator Pitch
Write a 2-3 sentence pitch that answers:
- What is it? (genre + platform)
- What do you do? (core verb / loop)
- Why is it special? (unique hook)
Template:
[Title] is a [genre] game for [platform] where you [core action]. Unlike [comparable game], [unique differentiator]. The game targets [audience] and can be built by [team size] in [timeframe].
Step 4: Core Loop Sketch
Map the 30-second, 5-minute, and session-level loops:
30-second loop (moment-to-moment):
→ [Input action] → [Feedback] → [Micro-decision]
5-minute loop (encounter/level):
→ [Setup] → [Challenge] → [Resolution] → [Reward]
Session loop (meta-progression):
→ [Goal selection] → [Attempt] → [Outcome] → [Unlock/Progress]
Step 5: Competitive Landscape
Identify 3-5 reference games and map where this concept sits:
Complex
│
Dwarf │ This Game
Fortress │ ★
│
Niche ───────────────┼─────────────── Mainstream
│
Obscure │ Stardew
Jam Game │ Valley
│
Simple
Step 6: Risk & Feasibility Check
| Risk | Likelihood | Mitigation |
|---|---|---|
| Scope creep | High | Define MVP feature set now |
| Tech unknown | Medium | Prototype risky mechanic first |
| Market saturation | Varies | Identify unique angle |
| Art bottleneck | High (solo) | Choose stylized/lo-fi aesthetic |
Output
Save the concept brief to artifacts/game-design/[project-name]-concept.md with:
- Title and elevator pitch
- Design pillars
- Core loop diagram
- Target platform, audience, and scope
- Competitive positioning
- Top 3 risks
Guidelines
- Quantity before quality in ideation — generate 5+ ideas before narrowing
- The verb is the game — if you can't describe the core action in one verb, keep refining
- Don't design the whole game here — that's what
/game-design-documentis for - Kill ideas that excite the designer but bore the player
- Scope is king for indie/solo devs — a finished small game beats an abandoned epic