| name | game-concept-generator |
| description | Generate game concepts through interactive constraint gathering. Use this when the user wants game ideas, needs inspiration, or wants to explore different game concepts based on their constraints. |
| domain | design |
| type | generator |
| version | 1.0.0 |
| allowed-tools | ["Write"] |
Game Concept Generator Skill
This skill generates game concepts through an interactive questioning process to understand user constraints, then produces 3-5 viable game ideas with detailed core loop descriptions.
When to Use This Skill
Invoke this skill when the user:
- Asks for "game ideas" or "game concepts"
- Says "I want to make a game but don't know what"
- Requests inspiration for a game jam or prototype
- Wants to explore what's possible within constraints
- Says "generate a [genre] game concept"
- Asks "what game should I build?"
Core Principle
Interactive constraint gathering ensures generated concepts are:
- ✅ Realistic for the user's constraints
- ✅ Aligned with their interests and skills
- ✅ Technically feasible with their tools
- ✅ Scoped appropriately for their timeline
- ✅ Unique and interesting to build
Interactive Questioning Workflow
Phase 1: Initial Context (Required)
Always start by asking these core questions:
I'll help you generate game concepts! Let me understand your constraints:
**1. Development Timeline**
How much time do you have?
- [ ] Weekend (2-3 days)
- [ ] Week (5-7 days)
- [ ] Month (4 weeks)
- [ ] Longer-term project
- [ ] Flexible / exploring ideas
**2. Target Platform**
Where will this game run?
- [ ] PC (Windows/Mac/Linux)
- [ ] Web (HTML5/WebGL)
- [ ] Mobile (iOS/Android)
- [ ] Console
- [ ] Multiple platforms
**3. Team Size**
Who's working on this?
- [ ] Solo developer
- [ ] 2-3 people
- [ ] Small team (4-6)
- [ ] Larger team
**4. Engine/Tools**
What are you building with?
- [ ] Godot
- [ ] Unity
- [ ] Unreal
- [ ] Custom engine
- [ ] Other: _______
Wait for user response before proceeding.
Phase 2: Genre & Mechanics (Clarification)
After receiving Phase 1 answers, ask:
Great! Now let's narrow down the type of game:
**5. Genre Preference**
Any genres you're excited about? (Select all that apply)
- [ ] Action (combat, reflexes, timing)
- [ ] Puzzle (logic, problem-solving)
- [ ] Strategy (planning, resource management)
- [ ] RPG (progression, stats, narrative)
- [ ] Roguelike (permadeath, procedural, runs)
- [ ] Platformer (jumping, movement)
- [ ] Simulation (systems, sandbox)
- [ ] Rhythm (music, timing)
- [ ] Tower Defense
- [ ] Auto-battler
- [ ] Narrative/Story
- [ ] Hybrid (mix of above)
- [ ] Open to anything
**6. Core Mechanic Interest**
What sounds fun to implement?
- [ ] Combat system
- [ ] Movement mechanics (physics, parkour)
- [ ] Puzzle mechanics
- [ ] Procedural generation
- [ ] AI/pathfinding
- [ ] Building/crafting
- [ ] Economy/trading
- [ ] Dialogue/narrative
- [ ] No preference - surprise me!
**7. Visual Complexity**
What art style are you comfortable with?
- [ ] Minimal (shapes, colors, no art needed)
- [ ] Pixel art (sprites, tiles)
- [ ] 3D simple (basic models)
- [ ] Hand-drawn 2D
- [ ] Vector graphics
- [ ] Text-based
Wait for user response before proceeding.
Phase 3: Scope & Experience (Refinement)
After receiving Phase 2 answers, ask:
Almost there! A few more questions to refine the concepts:
**8. Experience Level**
How comfortable are you with game dev?
- [ ] Beginner (first few games)
- [ ] Intermediate (made 2-5 games)
- [ ] Advanced (experienced developer)
- [ ] Expert (shipped multiple games)
**9. Scope Preference**
What kind of scope appeals to you?
- [ ] Tiny (one core mechanic, super focused)
- [ ] Small (2-3 mechanics, clear goal)
- [ ] Medium (multiple systems, 30-60 min gameplay)
- [ ] Ambitious (full game, hours of content)
**10. Inspiration or Constraints**
Anything else I should know?
- Favorite games to reference?
- Specific theme or setting?
- Technical limitations?
- Game jam theme?
- Must-have features?
(This is optional - just share anything relevant)
Wait for user response before proceeding.
Phase 4: Generate Concepts
After gathering all constraints, generate 3-5 game concepts.
Each concept must include:
- Concept Name (catchy, memorable)
- One-Line Pitch (elevator pitch)
- Core Loop (what the player does minute-to-minute)
- Key Mechanics (2-4 primary mechanics)
- Unique Hook (what makes this interesting)
- Scope Assessment (realistic timeline estimate)
- Technical Complexity (Low/Medium/High)
- Why This Works (how it fits their constraints)
Output Format
# Game Concepts for [User]
**Your Constraints:**
- Timeline: [X days/weeks]
- Platform: [Platform]
- Team: [Team size]
- Engine: [Engine]
- Genre interests: [Genres]
- Experience: [Level]
- Scope: [Preference]
---
## Concept 1: [Concept Name]
**One-Line Pitch:**
[Compelling elevator pitch in one sentence]
**Core Loop:**
- [Player action 1]
- [System response]
- [Player decision]
- [Outcome/reward]
→ Repeat with escalation
**Key Mechanics:**
- **[Mechanic 1]**: [Brief description]
- **[Mechanic 2]**: [Brief description]
- **[Mechanic 3]**: [Brief description]
**Unique Hook:**
[What makes this concept stand out? Why is it interesting?]
**Scope Assessment:**
- **Timeline:** [X days/weeks] - [Realistic/Ambitious/Tight]
- **MVP Features:** [List 3-4 must-have features for minimum viable version]
- **Stretch Goals:** [Optional features if time permits]
**Technical Complexity:** 🟢 Low | 🟡 Medium | 🔴 High
**Why This Works for You:**
[Explain how this concept aligns with their constraints, interests, and experience level]
**Similar Games for Reference:**
- [Game 1] (for mechanic X)
- [Game 2] (for feel/aesthetic)
---
## Concept 2: [Concept Name]
[Same structure as above]
---
## Concept 3: [Concept Name]
[Same structure as above]
---
## Concept 4: [Concept Name] (Wildcard)
[Same structure - this should be a more experimental/unique idea]
---
## Concept 5: [Concept Name] (Ambitious)
[Same structure - this should be a more ambitious take if they want a challenge]
---
## Recommendation
**Best fit for your constraints:**
[Concept #X] - [Name]
**Why:** [Explain why this concept is the best match for their specific situation]
**Next Steps:**
1. Pick a concept (or hybrid ideas)
2. Create a prototype roadmap
3. Start with core loop implementation
4. Iterate based on feel
**Questions to Ask Yourself:**
- Does this concept excite you enough to finish it?
- Can you describe the core loop in one sentence?
- What's the "aha!" moment for players?
- Is the scope realistic for your timeline?
---
Would you like me to:
- [ ] Elaborate on any concept
- [ ] Generate variations of a concept
- [ ] Create a prototype roadmap for a concept
- [ ] Combine ideas into a hybrid concept
Concept Generation Guidelines
Match Constraints Precisely
Timeline-Based Scoping:
-
Weekend (2-3 days):
- ONE core mechanic only
- Minimal art (shapes, colors)
- No complex systems
- Clear win/lose condition
- Example: "Dodge falling objects while collecting gems"
-
Week (5-7 days):
- 2-3 mechanics
- Simple pixel art or basic 3D
- One progression system
- 10-15 minutes of gameplay
- Example: "Roguelike with 3 weapons and 5 enemy types"
-
Month (4 weeks):
- 4-6 mechanics
- Multiple systems integration
- Progression and unlocks
- 30-60 minutes of content
- Example: "Tower defense with building + resource management"
Genre-Specific Patterns
Action Games:
- Core loop: Move → Shoot → Dodge → Survive
- Key: Responsive controls, satisfying feedback
- Prototyping focus: Feel and game juice
Puzzle Games:
- Core loop: Analyze → Plan → Execute → Solve
- Key: Clear rules, escalating complexity
- Prototyping focus: Puzzle generator and validation
Roguelikes:
- Core loop: Fight → Collect → Upgrade → Die → Retry
- Key: Run variety, meta-progression
- Prototyping focus: Procedural generation and balance
Strategy Games:
- Core loop: Gather → Build → Plan → Execute
- Key: Meaningful choices, visible consequences
- Prototyping focus: AI and resource balance
Auto-battlers:
- Core loop: Position → Watch → Adjust → Repeat
- Key: Strategic depth in preparation
- Prototyping focus: Combat simulation and balance
Platform-Specific Considerations
PC:
- Can use keyboard/mouse for complex controls
- Players expect depth and precision
- Long play sessions acceptable
Web:
- Must load quickly (<5MB)
- Simple controls (mouse or arrow keys)
- Short play sessions (5-10 min)
- Shareability is a feature
Mobile:
- Touch controls only (swipes, taps)
- Portrait or landscape (choose one)
- Short play sessions (1-3 min)
- Minimize text (visual communication)
Technical Complexity Ratings
🟢 Low Complexity:
- Built-in Godot nodes (CharacterBody2D, Area2D)
- Basic movement and collision
- Simple state machines
- No networking, no saving
🟡 Medium Complexity:
- Custom physics or pathfinding
- Procedural generation
- Save/load systems
- Multiple interconnected systems
🔴 High Complexity:
- Multiplayer/networking
- Advanced AI (behavior trees)
- Complex procedural generation
- 3D with custom shaders
Concept Quality Criteria
Every Concept Must Be:
- Achievable - Realistic within stated timeline
- Focused - Clear core loop, not trying to do everything
- Interesting - Has a unique hook or twist
- Testable - Can prototype core loop in 20% of timeline
- Expandable - Can add features incrementally
Red Flags to Avoid:
❌ Feature Creep Concepts
- "Open world RPG with crafting, building, and multiplayer"
- Too many systems for timeline
❌ Vague Core Loops
- "Explore and discover things"
- No clear minute-to-minute gameplay
❌ Technical Overreach
- "MMO" for a weekend project
- Requires tech beyond user's experience
❌ Derivative Without Twist
- "It's just like [popular game]"
- No unique selling point
❌ Scope Mismatch
- Tiny concept for month-long timeline
- Massive concept for weekend jam
Example Interaction
User: "I want to make a game but don't know what"
Skill:
I'll help you generate game concepts! Let me understand your constraints:
1. **Development Timeline** - How much time do you have?
2. **Target Platform** - Where will this game run?
3. **Team Size** - Who's working on this?
4. **Engine/Tools** - What are you building with?
User: "Weekend jam, web platform, solo, Godot"
Skill:
Great! A weekend web game in Godot - super achievable. Now:
5. **Genre Preference** - Action? Puzzle? Roguelike? (Pick favorites)
6. **Core Mechanic Interest** - What sounds fun to implement?
7. **Visual Complexity** - Minimal shapes? Pixel art? Text?
User: "Action or roguelike, love movement mechanics, minimal art is fine"
Skill:
Perfect! Almost there:
8. **Experience Level** - How comfortable are you with game dev?
9. **Scope Preference** - Tiny and focused? Or pack in features?
10. **Inspiration** - Any favorite games or themes?
User: "Intermediate, prefer focused, love Vampire Survivors"
Skill: [Generates 5 concepts, all fitting constraints]
Follow-Up Actions
After generating concepts, offer:
Elaboration
If user says "tell me more about Concept 2":
- Expand on implementation details
- Suggest art/audio references
- Break down development phases
- Estimate feature complexity
Variation
If user says "I like Concept 1 but [change]":
- Generate 3 variations with their modification
- Explain trade-offs of each variation
- Suggest hybrid approaches
Prototype Roadmap
If user says "I want to build Concept 3":
- Invoke
vertical-slice-planner skill
- Create day-by-day implementation plan
- Identify critical path features
- Suggest testing milestones
Hybrid Concepts
If user says "combine Concept 1 and 4":
- Analyze which mechanics blend well
- Create unified core loop
- Assess combined scope
- Suggest integration approach
Important Notes
- Interactive is key: ALWAYS ask questions before generating
- Constraints matter: A weekend concept ≠ month concept
- Realistic scoping: Better to under-promise and over-deliver
- Unique hooks: Every concept needs something interesting
- Reference games: Help user visualize with familiar examples
- Prototype-first: Concepts should be testable quickly
Example Invocations
User: "Generate a game idea for me"
User: "I want to make a roguelike for a weekend"
User: "Game jam tomorrow, need concepts"
User: "Help me brainstorm game ideas"
User: "What game should I build in Godot?"
Workflow Summary
- Ask Phase 1 questions (timeline, platform, team, tools)
- Wait for user response
- Ask Phase 2 questions (genre, mechanics, art)
- Wait for user response
- Ask Phase 3 questions (experience, scope, inspiration)
- Wait for user response
- Generate 3-5 concepts matching all constraints
- Offer follow-up actions (elaborate, vary, roadmap, hybrid)
- Create a markdown document capturing the game ideas
This iterative approach ensures concepts are tailored to the user's actual situation, not generic ideas.