Feature Discovery

Feature Discovery and Planning

Overview

This skill guides systematic feature discovery and planning to ensure requirements are well-understood before implementation begins. It's specifically designed for planning game features in the TripleDerby horse racing simulation.

When to Use

• Planning a new game feature or capability
• Analyzing feature requirements and scope
• Breaking down complex features into implementable tasks
• Conducting feasibility studies for new mechanics
• Designing game systems (race mechanics, betting, training, etc.)
• Scope planning for development iterations

Instructions

Phase 1: Discovery

1. Gather Context: Ask clarifying questions about feature goals, user experience, and game design intent
2. Identify Requirements: Document what the feature must do (functional requirements)
3. Explore Constraints: Understand technical, gameplay, and performance constraints
4. Map Use Cases: Identify player interactions and game scenarios
5. Review Codebase: Examine existing patterns and systems that may be affected or leveraged

Phase 2: Technical Analysis

1. Evaluate Feasibility: Assess technical viability within the current architecture
2. Identify Dependencies: Map relationships to existing systems (Race, Horse, Betting, etc.)
3. Review Data Model: Consider database schema changes needed
4. Assess Integration Points: Identify where new code connects to existing systems
5. Consider Performance: Evaluate impact on race simulation, UI responsiveness, etc.
6. Security & Validation: Identify data validation and security requirements

Phase 3: Planning

1. Break Down Tasks: Decompose the feature into concrete, implementable tasks
2. Sequence Work: Identify logical implementation phases and dependencies
3. Estimate Complexity: Note particularly complex or risky areas
4. Create Task List: Use TodoWrite to create a structured task breakdown
5. Define Milestones: Identify testable checkpoints and deliverables

Phase 4: Documentation

1. Write Specification: Create clear feature requirements document
2. Document Assumptions: Record decisions and assumptions made during planning
3. Define Success Criteria: Establish how to validate the feature works correctly
4. Create Implementation Plan: Write step-by-step implementation approach
5. Reference Templates: Use PLANNING_TEMPLATE.md and DISCOVERY_CHECKLIST.md

Examples

Example 1: New Game System

User: "I want to add a horse training system where players can improve their horses' stats over time"

Claude will:

• Ask about training mechanics (time-based? resource-based? mini-games?)
• Explore stat progression (linear? diminishing returns? caps?)
• Review existing Horse entity and stats structure
• Identify UI requirements for training interface
• Plan database schema changes for tracking training progress
• Break down into tasks: data model, service layer, UI, validation
• Create implementation roadmap with phases

Example 2: Race Feature Enhancement

User: "Design a weather system that affects race outcomes"

Claude will:

• Clarify weather types and their effects on racing
• Review RaceService and race simulation logic
• Identify where weather modifiers integrate into calculations
• Plan random weather generation vs player-controlled scenarios
• Consider UI for displaying weather conditions
• Document edge cases (weather changes mid-race?)
• Create phased implementation plan

Example 3: Requirements Analysis

User: "Help me plan a multiplayer betting feature where players can bet on each other's races"

Claude will:

• Ask about betting rules, stake limits, payout calculations
• Explore multiplayer architecture requirements
• Review existing betting system (if any)
• Identify security concerns (preventing cheating, validation)
• Plan real-time updates and notifications
• Break into phases: single-player betting, then multiplayer integration
• Create comprehensive task breakdown

Best Practices

Discovery Phase

• Ask open-ended questions to understand the "why" behind features
• Reference similar features in the existing codebase
• Consider player experience and game balance
• Document edge cases and unusual scenarios

Analysis Phase

• Read relevant existing code before proposing changes
• Identify patterns used elsewhere in the codebase (services, entities, repositories)
• Consider both happy path and error scenarios
• Think about testability

Planning Phase

• Break features into vertical slices when possible (end-to-end functionality)
• Sequence tasks to deliver testable increments
• Flag high-risk or complex areas that need extra attention
• Create detailed but flexible plans

Documentation Phase

• Write specifications that developers can implement from
• Include code examples or pseudocode for complex logic
• Reference existing patterns to follow
• Document open questions and areas needing future decisions

Output Format

After conducting discovery and planning, create a feature specification document in /docs/features/ that includes:

1. Feature Summary: One-paragraph overview
2. Requirements: Functional and non-functional requirements
3. Technical Approach: Architecture and integration points
4. Implementation Plan: Phased task breakdown
5. Success Criteria: How to validate correctness
6. Open Questions: Items needing further clarification

Use the PLANNING_TEMPLATE.md as a starting point for documentation.

Output Location: All feature specification documents should be saved to /docs/features/[feature-name].md

Integration with Existing Systems

When planning features for TripleDerby, consider integration with:

• Core Entities: Horse, Race, RaceRun, RaceRunHorse, Player, etc.
• Services: RaceService, simulation logic, game flow
• Data Layer: Entity Framework, repositories, ModelBuilderExtensions
• UI: Game interface, player controls, visual feedback
• Game Balance: Stats, probabilities, progression systems

Notes

• This skill uses Read, Grep, and Glob tools to explore the codebase during discovery
• Creates task lists using TodoWrite for structured planning
• May use AskUserQuestion to clarify ambiguous requirements
• Produces markdown documentation for feature specifications
• Focuses on practical, implementable plans rather than theoretical designs

CRITICAL GIT WORKFLOW RULES

NEVER commit or push code without EXPLICIT user approval.

When implementing features:

1. Discovery Phase: Research, design, document - NO git commands
2. Implementation Phase: Write code, run tests - NO git commits
3. WAIT for user approval: User must explicitly say "commit this" or "push this"
4. ONLY THEN: Create commits and push to remote

Examples of what NOT to do:

• ❌ "Let me commit Phase 1" (without asking first)
• ❌ Automatically committing after tests pass
• ❌ Assuming user wants code committed

Correct workflow:

• ✅ Implement feature
• ✅ Run tests to verify
• ✅ Report results to user
• ✅ ASK: "Would you like me to commit these changes?"
• ✅ WAIT for explicit approval
• ✅ ONLY THEN run git commands

COMMIT AND PR MESSAGE RULES

NEVER include Claude Code attribution or self-reference in commits or PRs.

When creating commits or pull requests:

• ❌ NEVER add "🤖 Generated with [Claude Code]" footer
• ❌ NEVER add "Co-Authored-By: Claude Sonnet" tag
• ❌ NEVER mention Claude, AI, or automated generation
• ✅ Write commit messages as if written by the developer
• ✅ Focus on what changed and why
• ✅ Use professional, technical language only

Examples:

BAD - DO NOT DO THIS:

Feature implementation complete

This adds the new commentary system.

🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

GOOD - DO THIS:

Feature implementation complete

This adds the new commentary system with event detection
and natural language variation.