- Home
- Skills
- Game Development
- Game Design Session: Idea to Design Doc
Game Design Session: Idea to Design Doc
An interviewer that turns a game idea into a living design doc and a first milestone, in your words.
$5
Works with the AI Agents you already use
Game Design Session: Idea to Design Doc
Example session with this skill installed
I want to design a 2D top-down extraction shooter for mobile. Let's start with the Core Loop and then talk about the Tech Stack.
- Read your context and instructions
- Compiled the game design session
I have initialized the design session. I've created the GDD and recorded the initial mobile extraction shooter pitch.
Game Design Update: Extraction Shooter (Mobile)
Decisions Logged:
- Genre: 2D Top-down Extraction Shooter.
- Platform: Mobile (Primary).
- Session Goal: Define Core Loop and Engine.
1. Core Loop (Round 2)
- Moment-to-moment: Scavenging crates, tactical positioning with touch controls, and high-tension fire fights.
- Per session: 5-minute raids, inventory management, and "Extract or Die" decision points.
- Long term: Base upgrades, weapon mastery, and seasonal leaderboards based on loot value.
2. Tech Stack Comparison
| Engine | Recommendation | Reason |
|---|---|---|
| Godot | Recommended | Excellent 2D performance, small binary size for mobile, and high agent-compatibility for automated testing. |
| Unity | Optional | Strongest ad-network integration but heavier overhead for simple 2D. |
| PlayCanvas | Optional | Best for zero-install browser play, but lacks the robust native mobile performance of Godot. |
3. Open Questions for Design
- Will the map be procedurally generated or hand-crafted?
- What is the penalty for death? (Full loot drop, partial, or currency loss)
- Should we support portrait or landscape orientation for mobile?
- Do you want a persistent player hub between raids?
Next steps
- Pick one of the Core Loop options or provide your own.
- Confirm the engine choice to unlock the "Scope and Platforms" round.
- Define the death penalty to move the Challenge section to DECIDED.
Connects securely to your tools. The creator never sees your data.
About this skill
Game design stalls when you lose track of decisions or fill gaps with placeholder ideas that break the core loop later. This skill acts as a structured design partner that interviews you through every layer of production, ensuring your vision is documented without the agent overstepping its bounds.
What it does
- Structured Interviews runs focused rounds covering pitch, core loop, progression, and monetization to extract your vision.
- Dynamic Documentation maintains a Game Design Document (GDD) with sections clearly marked as OPEN, RECOMMENDED, or DECIDED.
- Evidence-Based Options provides researched trade-offs, comparable game teardowns, and economy models only when requested.
- Tech Stack Advisor compares engines like Godot, Three.js, or Unreal based on platform targets and agent-compatibility for building.
- Milestone Planning translates high-level decisions into a technical task list for a "First Playable" vertical slice.
How it works
- Context Loading scans your
docs/design/GDD.mdanddocs/DECISIONS.mdor creates them from templates if missing. - Round-Based Debriefing asks up to 4 targeted questions per session, prioritizing your energy while keeping the design cohesive.
- Research Spikes performs deep dives into comparable games or legal/store requirements for specific platforms when you hit a wall.
- Decision Logging records every choice with a date and rationale so you never forget why a mechanic was changed.
- Execution Prep converts the finalized design into an actionable milestone plan with technical dependencies.
Frameworks & tools
Works with any engine including Godot, Unity, Unreal, and web-based stacks like Three.js or Babylon.js. It assumes a file-based documentation structure in Markdown.
Why this beats prompting it yourself
Generic prompts often "hallucinate" game features you didn't ask for, leading to scope creep and loss of creative control. This skill is strictly reactive, marking gaps as OPEN rather than inventing them, and forces technical rigor by linking design choices directly to engine capabilities.
Use cases
- Drafting a comprehensive GDD from a one-sentence "X meets Y" elevator pitch.
- Solving a specific mechanical bottleneck like economy inflation or multiplayer persistence.
- Evaluating whether a project should be built in Roblox vs. a custom Three.js engine.
- Creating a scoped-down milestone plan for a weekend game jam or a vertical slice.
Known limitations
Does not provide final legal advice or write the actual game code during the design session. Requires manual confirmation of recommendations to move sections from OPEN to DECIDED.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Game Development.
Will not invent design elements without developer input. Does not provide final legal advice. Requires Markdown templates for GDD and Decisions log.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago