Read a prototype GDD and generate a phased implementation roadmap with concrete tasks, organized by implementation phases with specific deliverables and test criteria.
Installation
Details
Usage
After installing, this skill will be available to your AI coding assistant.
Verify installation:
npx agent-skills-cli listSkill Instructions
name: prototype-roadmap-planner description: Read a prototype GDD and generate a phased implementation roadmap with concrete tasks, organized by implementation phases with specific deliverables and test criteria. domain: project type: planner version: 1.0.0 allowed-tools:
- Read
- Write
- Glob
Prototype Roadmap Planner Skill
This skill reads a prototype GDD and generates a detailed implementation roadmap. It extracts implementation phases from the GDD, breaks down each phase into concrete tasks, and creates a day-by-day or hour-by-hour development plan for rapid prototyping.
When to Use This Skill
Invoke this skill when the user:
- Says "create roadmap for prototype"
- Asks "plan implementation for prototype"
- Wants to break down prototype into actionable tasks
- Has a prototype GDD and needs a development plan
- Says "what's my implementation plan?" or "what should I build first?"
- Wants a day-by-day breakdown of prototype work
Core Principle
Roadmaps turn design into action:
- ✅ Extract implementation phases from prototype GDD
- ✅ Break each phase into concrete, testable tasks
- ✅ Create hour-by-hour or day-by-day timeline
- ✅ Define clear deliverables and test criteria for each phase
- ✅ Prioritize tasks to get playable core loop as fast as possible
- ✅ Roadmap is actionable, not abstract
Visible-First Ordering Philosophy
Build what you can SEE before what you can't.
Traditional development often goes: Data → Systems → Features → UI. For prototypes, invert this:
VISIBLE-FIRST ORDER:
1. Visual elements (sprites, scenes, UI) ← See it first
2. Player interaction (input, feedback) ← Feel it second
3. Core systems (logic, rules) ← Make it work
4. Data structures (resources, configs) ← Refine later
5. Polish (juice, effects) ← Last
Why Visible-First?
- Observable progress: Seeing sprites move on screen motivates more than seeing console logs
- Early feedback: Spot design problems when you can see the game, not just read data
- Testable sooner: Playtest with placeholder logic, not placeholder visuals
- Avoid over-engineering: Build only the systems you actually need for what you see
Example - Grid Combat:
❌ Traditional Order (invisible progress):
- Create grid data structure
- Create pathfinding algorithm
- Create unit stats system
- Create UI to display it all
- Finally see something
✅ Visible-First Order (observable progress):
- Create grid visual (TileMap or sprites) ← See the board
- Create unit sprites that can be placed ← See the pieces
- Create click-to-select, click-to-move ← Feel interaction
- Add pathfinding for valid moves ← Make it smart
- Add stats and damage calculation ← Make it meaningful
At each step, you can SEE and FEEL progress.
Ordering Rules
- Scenes before scripts: Create the visual scene structure first, add behavior second
- Sprites before data: Use placeholder sprites immediately, define stat Resources later
- Hardcode before configure: Hardcode values to get it working, extract to data files later
- Input before logic: Get player clicking/pressing first, add game rules second
- Feedback before polish: Show that something happened (even ugly), make it pretty later
Dependency Mapping (Visible-First)
When mapping task dependencies, follow this order:
Phase 1: Core Visuals
├── Scene structure (Main.tscn, Game.tscn)
├── Player/unit sprites visible
├── Basic camera/viewport
└── Placeholder UI elements
Phase 2: Core Interaction
├── Input handling (click, WASD)
├── Visual feedback (selection, movement)
├── State changes visible (health bars, etc.)
└── Basic sound effects (optional)
Phase 3: Core Systems
├── Game logic (turn order, combat rules)
├── Data integration (stats affect outcomes)
├── Win/lose conditions
└── State management
Phase 4: Data & Balance
├── Extract hardcoded values to Resources
├── Create data files (JSON, .tres)
├── Balance pass (tune numbers)
└── Configuration systems
Phase 5: Polish
├── Animations and transitions
├── Juice (screenshake, particles)
├── UI polish
└── Audio polish
Roadmap Output Structure
The skill generates a detailed, task-oriented roadmap:
# [GAME TITLE] - Prototype Implementation Roadmap
**Based On:** Prototype GDD v[X.Y]
**Timeline:** [X days from GDD]
**Target:** [Core question from GDD]
**Created:** [Date]
---
## Quick Start
**Critical Path to Playable Core Loop:**
1. [Most important task from Phase 1]
2. [Most important task from Phase 1]
3. [First test milestone - "can play one round"]
**Estimated time to playable:** [X hours]
---
## Phase 1: [Phase Name] ([Timeline] - [X hours])
**Goal:** [What you're building - from GDD]
**Deliverables:**
- [Deliverable 1 from GDD]
- [Deliverable 2 from GDD]
**Test Criteria:** [How you know it works - from GDD]
### Tasks
#### Task 1.1: [Clear Action Verb] [Specific Thing]
**Time:** [X hours]
**What:** [One sentence describing the visible/testable outcome]
**How:**
- [Implementation step 1]
- [Implementation step 2]
- [Implementation step 3]
**Acceptance:**
- [ ] [Specific testable criterion - observable]
- [ ] [Specific testable criterion - observable]
**Files:**
- `[file path]` - [purpose]
- `[file path]` - [purpose]
**Hardcoded Values:**
- [Value name]: [value] (e.g., movement_speed: 200)
- [Value name]: [value]
**Dependencies:** None / [Task X.Y must be complete]
---
#### Task 1.2: [Clear Action Verb] [Specific Thing]
**Time:** [X hours]
**What:** [One sentence describing the visible/testable outcome]
**How:**
- [Implementation step 1]
- [Implementation step 2]
**Acceptance:**
- [ ] [Specific testable criterion]
- [ ] [Specific testable criterion]
**Files:**
- `[file path]` - [purpose]
**Hardcoded Values:**
- [Value name]: [value]
**Dependencies:** [Task 1.1]
---
[Continue for all tasks in phase]
**Phase 1 Checkpoint:**
- [ ] All deliverables complete
- [ ] Test criteria passed: [test description]
- [ ] Ready to proceed to Phase 2
---
## Phase 2: [Phase Name] ([Timeline] - [X hours])
**Goal:** [What you're building - from GDD]
**Deliverables:**
- [Deliverable 1 from GDD]
- [Deliverable 2 from GDD]
**Test Criteria:** [How you know it works - from GDD]
### Tasks
[Same structure as Phase 1]
**Phase 2 Checkpoint:**
- [ ] All deliverables complete
- [ ] Test criteria passed: [test description]
- [ ] Ready to proceed to Phase 3
---
[Continue for all phases from GDD]
---
## Testing & Validation
**After Each Phase:**
- [ ] Run test criteria from phase
- [ ] Document any issues/blockers
- [ ] Adjust timeline if behind schedule
**Final Prototype Test (End of Timeline):**
- [ ] Can complete full playthrough
- [ ] Critical questions from GDD answerable
- [ ] Playtesters can understand core loop
- [ ] Success metrics measurable
---
## Risk Mitigation Plan
**From GDD Risk Section:**
**Risk:** [Risk from GDD]
**If This Happens:** [Fallback from GDD]
**Action Items:**
- [Specific task to mitigate]
- [Specific task to implement fallback]
[Repeat for each risk from GDD]
---
## Daily Schedule (for [X]-day timeline)
### Day 1: [Date]
**Focus:** [Phase name]
- Morning (4h): [Tasks]
- Afternoon (4h): [Tasks]
**End of Day Goal:** [Deliverable]
### Day 2: [Date]
**Focus:** [Phase name]
- Morning (4h): [Tasks]
- Afternoon (4h): [Tasks]
**End of Day Goal:** [Deliverable]
[Continue for each day]
---
## Content Checklist
**Assets to Create:**
- [ ] [Asset 1 from GDD content scope]
- [ ] [Asset 2 from GDD content scope]
**Data/Balance:**
- [ ] [Stats for units/enemies from GDD]
- [ ] [Configuration values]
---
**END OF ROADMAP**
**Next Steps:**
1. Review this roadmap
2. Set up development environment
3. Start Phase 1, Task 1.1
4. Test frequently, adjust timeline as needed
Workflow
Step 1: Read Prototype GDD
-
Find the prototype GDD:
- Look for
docs/*prototype-gdd*.mdor ask user for path - Read the entire GDD
- Look for
-
Extract key information:
- Timeline (weekend/week/month from GDD)
- Critical questions being tested
- Implementation phases (from "Implementation Phases" section)
- Each phase's goal, deliverables, test criteria
- Content scope (units, mechanics, etc.)
- Risk mitigation plans
- Success metrics
Step 2: Break Down Each Phase into Tasks
For each implementation phase in the GDD:
-
Extract phase metadata:
- Phase name and timeline (e.g., "Day 1, Morning - 4h")
- Goal statement
- Deliverables list
- Test criteria
-
Infer concrete tasks from deliverables:
- Each deliverable becomes 1-3 tasks
- Tasks should be 1-3 hours each (small enough to track progress)
- Tasks should have clear acceptance criteria
Example: GDD Deliverable: "WASD movement, one weapon, enemy spawning"
Becomes tasks:
- Task 1.1: Player Character with WASD Movement (2h)
- Task 1.2: Basic Weapon Auto-Fire System (1.5h)
- Task 1.3: Enemy Spawner and Basic Enemy (1.5h)
- Define acceptance criteria:
- What specific behaviors must work
- How to test it
- Observable success conditions
Example: Task 1.1: Player Character with WASD Movement Acceptance Criteria:
- Player sprite appears in center of screen
- W/A/S/D keys move player in correct directions
- Movement speed is 200 px/s
- Player cannot move off-screen
-
Identify file paths:
- Where will code live (scripts/player/player.gd)
- What scenes to create (scenes/player/Player.tscn)
- Based on project structure and GDD organization
-
Map dependencies (Visible-First):
- Which tasks must complete before others
- Follow visible-first order: visuals → interaction → systems → data → polish
- Prioritize tasks that produce observable results early
Step 3: Create Critical Path
Identify the shortest path to playable core loop:
- Which tasks MUST complete to test the core mechanic?
- What's the absolute minimum to get "one round" working?
- Usually 3-5 tasks from Phase 1
Output as "Quick Start" section:
Critical Path to Playable Core Loop:
1. Player movement (Task 1.1)
2. Enemy spawning (Task 1.3)
3. Basic combat (Task 1.2)
Estimated time to playable: 5 hours
Step 4: Generate Daily Schedule
For short timelines (weekend/week):
- Break phases into day-by-day chunks
- Assign tasks to morning/afternoon blocks
- Set end-of-day goals (deliverables)
Example for weekend timeline:
Day 1:
- Morning: Phase 1 (Core Loop Foundation)
- Afternoon: Phase 2 (Economy System)
End of Day Goal: Can place units and earn gold
Day 2:
- Morning: Phase 3 (Full Content)
- Afternoon: Phase 4 (Wave Progression)
End of Day Goal: Full 10-wave game loop works
Day 3:
- All day: Phase 5 (Polish & Juice)
End of Day Goal: Prototype ready for playtesting
Step 5: Extract Content Checklist
From GDD's "Content Scope" section:
- List all units/enemies/items to create
- Extract stats and attributes
- Create checklist format
Example:
Content Checklist:
- [ ] Goblin (15g, 30 HP, 5 ATK)
- [ ] Skeleton (25g, 40 HP, 8 ATK)
- [ ] Warrior Hero (50 HP, 10 ATK, 20g drop)
Step 6: Map Risk Mitigation to Tasks
From GDD's "Risk Mitigation" section:
- Each risk becomes potential task
- Mitigation = proactive task (do during normal dev)
- Fallback = conditional task (do if risk occurs)
Example:
Risk: Auto-battle too fast/confusing
Mitigation Task: Add damage numbers and 1-2s combat pacing (Phase 5, Task 5.2)
Fallback Task: Add "Pause" button during combat (conditional - if playtest reveals issue)
Step 7: Save Roadmap
Save to: docs/[game-name]-prototype-roadmap.md
Output to user:
Created prototype roadmap with:
- [X] phases over [timeline]
- [X] concrete tasks
- Day-by-day schedule
- Critical path to playable in [X] hours
Next steps:
1. Review roadmap: docs/[game]-prototype-roadmap.md
2. Set up dev environment
3. Start Phase 1, Task 1.1
4. Test after each task, track progress
Example Interaction
User: "Create roadmap for prototype"
Skill:
Reading prototype GDD...
Found: docs/dungeon-investor-prototype-gdd.md
- Weekend timeline (2-3 days)
- 5 implementation phases defined
- Testing: Economic auto-battle dungeon defense
- Goal: Answer 5 critical questions
Breaking down phases into tasks...
Phase 1: Core Loop Foundation (Day 1 AM, 4-5h)
- 3 tasks (Player setup, Drag-drop, Auto-battle)
Phase 2: Economy System (Day 1 PM, 3-4h)
- 4 tasks (Gold UI, Shop, Cost system, Payouts)
Phase 3: Full Content (Day 2 AM, 4-5h)
- 3 tasks (5 monsters, 3 traps, 3 heroes)
Phase 4: Wave Progression (Day 2 PM, 3-4h)
- 4 tasks (Wave system, Scaling, Win/lose, Restart)
Phase 5: Polish & Juice (Day 3, 4-6h)
- 5 tasks (Damage numbers, Animations, UI clarity, etc.)
Generating daily schedule...
Day 1: Core Loop + Economy → playable with gold system
Day 2: Full Content + Waves → complete 10-wave loop
Day 3: Polish → ready for playtesting
Critical path to playable: 5 hours (3 tasks from Phase 1)
Roadmap saved to: docs/dungeon-investor-prototype-roadmap.md
Total: 19 tasks across 5 phases over 3 days
Ready to start development!
Integration with Other Skills
Reads From:
prototype-gdd-generator- Input GDD defining the prototype
Feeds Into:
changelog-updater- Track progress as tasks completesystems-bible-updater- Document systems as they're built
Works With:
version-control-helper- Manage commits per task/phasegdscript-quality-checker- Review code quality after each phase
Quality Checklist
Before finalizing roadmap:
Ordering & Structure:
- ✅ Follows visible-first ordering (visuals → interaction → systems → data → polish)
- ✅ Early tasks produce observable results (not invisible infrastructure)
- ✅ All phases from GDD are included
- ✅ Critical path identified (shortest route to playable)
Task Quality:
- ✅ Each task uses 5-part format (What, How, Acceptance, Files, Hardcoded Values)
- ✅ Tasks are 1-3 hours each (trackable progress)
- ✅ Each phase broken into concrete, testable tasks
- ✅ Detail level is actionable but not over-specified
- ✅ Hardcoded values listed (to extract to data later)
Coverage:
- ✅ Daily schedule created for timeline
- ✅ All deliverables from GDD mapped to tasks
- ✅ Test criteria from GDD included in checkpoints
- ✅ Content checklist extracted from GDD scope
- ✅ Risk mitigation mapped to tasks
Key Principles
DO:
- ✅ Visible-first ordering: Scenes → Interaction → Systems → Data → Polish
- ✅ Extract phases directly from GDD (don't invent new structure)
- ✅ Break deliverables into concrete, actionable tasks
- ✅ Include specific file paths and acceptance criteria
- ✅ Use 5-part task format: What, How, Acceptance, Files, Hardcoded Values
- ✅ Create testable checkpoints after each phase
- ✅ Identify critical path to playable core loop
- ✅ Map risks to mitigation tasks
- ✅ Hardcode values first, extract to Resources later
DON'T:
- ❌ Start with data structures before visuals (invisible progress)
- ❌ Build systems before you can see them working
- ❌ Add tasks not related to GDD deliverables
- ❌ Make tasks too large (>3 hours is too big)
- ❌ Skip test criteria or acceptance criteria
- ❌ Ignore dependencies between tasks
- ❌ Create vague tasks ("make it better")
- ❌ Over-specify tasks (save code details for implementation)
- ❌ Assume file structure (use project's actual structure)
Task Breakdown Guidelines
From GDD Deliverable to Tasks
GDD says: "WASD movement, one weapon, enemy spawning"
Becomes:
-
Task: Player Character Setup
- Create Player.tscn scene
- Create player.gd script with CharacterBody2D
- Implement WASD input handling
- Set movement speed to 200 px/s
- Test: Player moves in all 4 directions
-
Task: Basic Weapon System
- Create Weapon.gd script
- Implement auto-fire timer
- Spawn projectiles on timer tick
- Test: Weapon fires automatically every second
-
Task: Enemy Spawner
- Create Enemy.tscn scene
- Create spawner.gd script
- Spawn enemy every 3 seconds
- Test: Enemies appear and move toward player
Task Size Guidelines
Good task sizes:
- 1-3 hours of focused work
- Single responsibility (one system or feature)
- Testable in isolation
- Clear done criteria
Too large:
- "Implement combat system" (4-8 hours, multiple systems)
- Split into: damage calculation, health system, hit detection
Too small:
- "Add constant for move speed" (<15 min)
- Combine with larger task
Feature Description Detail Level
Each task in the roadmap needs enough detail to be actionable without ambiguity, but not so much that it becomes a full specification. Use this framework:
The 5-Part Task Description
Every task description should include these five elements:
#### Task X.Y: [Clear Action Verb] [Specific Thing]
**What:** One sentence describing the visible/testable outcome.
**How:** 3-5 bullet points of implementation steps.
**Acceptance:** 2-4 checkbox criteria that prove it works.
**Files:** List of files to create or modify.
**Hardcoded Values:** Any magic numbers to use (extract to data later).
Detail Level Examples
Too Vague (Unusable)
#### Task 1.1: Add player movement
Create player movement system.
❌ What kind of movement? Grid? Free? How fast? What input?
Too Detailed (Over-specified)
#### Task 1.1: Add player movement
Create a CharacterBody2D-based player controller using the
built-in input map system. The player should use a state
machine with Idle, Walking, and Running states. Movement
should use velocity-based physics with acceleration of
800 px/s² and deceleration of 1200 px/s². Maximum walk
speed is 150 px/s and run speed is 280 px/s when holding
Shift. Implement 8-directional movement with diagonal
normalization. Add dust particles when changing direction...
❌ This is a specification, not a roadmap task. Save this for implementation.
Just Right (Actionable)
#### Task 1.1: Player Grid Movement
**What:** Player character moves between grid tiles using arrow keys/WASD.
**How:**
- Create Player.tscn with Sprite2D and CollisionShape2D
- Add player.gd with input handling
- Move one tile per keypress (no diagonal for now)
- Animate tween between tiles (0.2s duration)
**Acceptance:**
- [ ] Player sprite visible on grid
- [ ] Arrow keys move player one tile in pressed direction
- [ ] Player cannot move outside grid boundaries
- [ ] Movement has smooth tween animation
**Files:**
- `scenes/player/Player.tscn`
- `scripts/player/player.gd`
**Hardcoded Values:**
- Grid size: 64x64 pixels per tile
- Movement duration: 0.2 seconds
- Grid bounds: 12x12 tiles
✅ Clear what to build, how to approach it, how to test it, specific enough to start immediately.
Detail Level by Task Type
Visual/Scene Tasks (High Visual Detail)
Focus on: What it looks like, where things are positioned, placeholder assets to use
#### Task 2.1: Combat Grid Display
**What:** 12x12 grid visible in center of screen with distinct tiles.
**How:**
- Create CombatGrid.tscn with TileMapLayer
- Use placeholder tile (64x64 colored squares)
- Light tiles for walkable, dark for obstacles
- Center grid in viewport (1920x1080)
**Acceptance:**
- [ ] Grid renders 12x12 tiles
- [ ] Grid is centered on screen
- [ ] Walkable vs obstacle tiles visually distinct
- [ ] Grid scale appropriate (fills ~60% of screen width)
**Hardcoded Values:**
- Tile size: 64x64 pixels
- Walkable color: #4a5568
- Obstacle color: #1a202c
Interaction Tasks (High Behavior Detail)
Focus on: What input does, what feedback player sees
#### Task 2.3: Unit Selection System
**What:** Click on friendly unit to select it, showing selection indicator.
**How:**
- Add Area2D with input detection to Unit.tscn
- Create selection_indicator.tscn (sprite or shader outline)
- Track currently_selected_unit in CombatManager
- Click empty tile or enemy = deselect
**Acceptance:**
- [ ] Clicking unit shows selection indicator
- [ ] Only one unit selected at a time
- [ ] Clicking elsewhere deselects
- [ ] Selected unit info appears in UI panel
**Hardcoded Values:**
- Selection indicator: Yellow outline, 2px
- Click detection radius: 32 pixels
System/Logic Tasks (High Rule Detail)
Focus on: Rules, formulas, edge cases
#### Task 3.2: Attack Damage Calculation
**What:** When unit attacks, calculate and apply damage using stats.
**How:**
- Create damage_calculator.gd utility
- Formula: damage = attack_power - (defense / 2)
- Minimum damage: 1 (always deal at least 1)
- Apply damage to target's current_hp
- Emit signal when unit dies (hp <= 0)
**Acceptance:**
- [ ] Attack reduces target HP by calculated amount
- [ ] Defense reduces incoming damage
- [ ] Minimum 1 damage dealt even vs high defense
- [ ] Unit removed from grid when HP reaches 0
**Hardcoded Values:**
- Defense divisor: 2
- Minimum damage: 1
- Test values: Attacker (ATK 10) vs Defender (DEF 4) = 8 damage
Data/Resource Tasks (High Structure Detail)
Focus on: Data shape, field names, example values
#### Task 4.1: Unit Stats Resource
**What:** Create Resource class to hold unit statistics.
**How:**
- Create UnitStats.gd extending Resource
- Export variables for: hp, attack, defense, move_range, attack_range
- Create 3 example .tres files (Soldier, Archer, Knight)
- Update Unit.tscn to use UnitStats resource
**Acceptance:**
- [ ] UnitStats resource class exists
- [ ] Can create .tres files with stats in editor
- [ ] Unit.tscn has exported UnitStats variable
- [ ] Changing .tres values affects unit behavior
**Example Data:**
- Soldier: HP 30, ATK 8, DEF 4, Move 3, Range 1
- Archer: HP 20, ATK 10, DEF 2, Move 3, Range 5
- Knight: HP 50, ATK 6, DEF 8, Move 2, Range 1
What NOT to Include in Task Descriptions
❌ Don't include:
- Full code implementations (that's for coding time)
- Design justifications (that's in the GDD)
- Alternative approaches (pick one, note alternatives in comments if needed)
- Polish details (animations, particles, sounds - save for polish phase)
- Edge cases you haven't thought through yet
✅ Do include:
- Enough to start coding immediately
- Clear done criteria
- Specific numbers (even if they'll change later)
- File locations
- Dependencies on other tasks
Example Output Structure
Example 1: Dungeon Investor (Visible-First Ordering)
Phase 1: Core Visuals & Scene (Day 1 Morning)
- Task 1.1: Dungeon Room Grid Display (1h) ← SEE the board first
- Task 1.2: Monster/Trap Placeholder Sprites (30min) ← SEE units
- Task 1.3: Hero Placeholder Sprites (30min) ← SEE enemies
- Task 1.4: Gold Counter UI Element (30min) ← SEE economy
Phase 2: Core Interaction (Day 1 Afternoon)
- Task 2.1: Drag-Drop Monster Placement (1.5h) ← FEEL placement
- Task 2.2: Shop Panel with Clickable Items (1h) ← FEEL purchasing
- Task 2.3: Hero Spawning and Walking (1h) ← SEE heroes enter
Phase 3: Core Systems (Day 2 Morning)
- Task 3.1: Basic Auto-Battle Combat (1.5h) ← Combat logic
- Task 3.2: Purchase/Refund Gold Logic (1h) ← Economy logic
- Task 3.3: Hero Death → Gold Earned (30min) ← Reward loop
- Task 3.4: Wave Counter System (1h) ← Progression
Phase 4: Content & Data (Day 2 Afternoon)
- Task 4.1: Monster Stats Resource Class (30min)
- Task 4.2: Implement 5 Monster Variants (1.5h)
- Task 4.3: Implement 3 Trap Variants (1h)
- Task 4.4: Hero Stat Scaling Per Wave (30min)
Phase 5: Polish (Day 3)
- Task 5.1: Damage Numbers Floating Text (1h)
- Task 5.2: Death Animations (1h)
- Task 5.3: UI Clarity Pass (1h)
- Task 5.4: Win/Lose Screens (1h)
Notice the ordering:
- Phase 1: Everything visual (you can SEE the game board)
- Phase 2: Everything interactive (you can CLICK and DRAG)
- Phase 3: Everything logical (it starts WORKING)
- Phase 4: Everything data-driven (it becomes BALANCED)
- Phase 5: Everything polished (it feels GOOD)
Example 2: Grid Combat (Visible-First Ordering)
Phase 1: Visual Foundation (4h)
- Task 1.1: Combat Grid TileMap (1h) ← SEE the battlefield
- Task 1.2: Unit Sprites on Grid (1h) ← SEE the pieces
- Task 1.3: HP Bars Above Units (1h) ← SEE health state
- Task 1.4: Turn Order UI Panel (1h) ← SEE who acts next
Phase 2: Interaction Layer (4h)
- Task 2.1: Click to Select Unit (1h) ← FEEL selection
- Task 2.2: Highlight Valid Move Tiles (1h) ← SEE options
- Task 2.3: Click to Move Unit (1h) ← FEEL movement
- Task 2.4: Click Enemy to Attack (1h) ← FEEL combat
Phase 3: Game Logic (4h)
- Task 3.1: Turn Order Calculation (1h)
- Task 3.2: Movement Range Validation (1h)
- Task 3.3: Attack Damage Calculation (1h)
- Task 3.4: Death and Victory Conditions (1h)
Phase 4: Data Extraction (2h)
- Task 4.1: UnitStats Resource Class (30min)
- Task 4.2: Create Unit .tres Files (30min)
- Task 4.3: Extract Hardcoded Values (30min)
- Task 4.4: Balance Pass (30min)
Phase 5: Polish (2h)
- Task 5.1: Movement Animation Tweens (30min)
- Task 5.2: Attack Animation (30min)
- Task 5.3: Damage Number Popups (30min)
- Task 5.4: Turn Transition Effects (30min)
Each task includes:
- Time estimate
- What/How/Acceptance criteria
- File paths
- Dependencies
- Hardcoded values
Timeline Formats
Weekend (2-3 days)
- Day-by-day breakdown
- Morning/afternoon blocks
- Hourly estimates
- End-of-day goals
Week (5-7 days)
- Day-by-day breakdown
- Daily focus areas
- Half-day blocks
- Milestone at midpoint and end
Two Weeks
- Week 1 / Week 2 breakdown
- Daily goals within weeks
- Weekly milestones
- Buffer time for iteration
Month
- Weekly breakdown
- Phase-level milestones
- More flexible daily structure
- Regular playtest checkpoints
This skill transforms a prototype GDD into an actionable, task-oriented roadmap with concrete steps, time estimates, and testable checkpoints. It's designed for rapid prototyping where speed and clarity are critical.
More by cautiouskurns
View allCreate and maintain a Game Design Bible documenting design vision, pillars, mechanics philosophy, and creative direction. Use this when establishing design principles or documenting high-level creative decisions.
Implement a feature spec by creating all necessary scenes, scripts, and resources. Use this when the user wants to implement a documented feature from the docs/features/ directory.
Analyze Godot scene files (.tscn) for structural issues, performance problems, and optimization opportunities. Use this when the user wants to optimize a scene, check scene structure, or troubleshoot scene-related issues.
Build and maintain a living architecture document tracking scene trees, data structures, system hierarchies, and game architecture. Use this when visualizing game structure, documenting scene hierarchies, or tracking data schemas.
