Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Game Testing

    1

    Systematic QA and bug-hunting for 2D/3D web games with severity-ranked reporting and logic verification.

    Free

    2 usesSecurity scanned

    About this skill

    Game development moves fast, and manual playtesting often misses edge cases in collision math or state transitions. This skill provides a systematic QA framework for 2D and 3D web games, focusing on Three.js and browser-based projects. It transforms vague "does it work?" checks into structured bug reports with evidence-backed findings.

    What it does

    • Rule verification maps game logic against RULES.md to identify spec gaps and logic errors.
    • Static code review scans for common pitfalls in game loops, input handling, and entity spawning.
    • Automated logic testing generates unit tests for pure functions like scoring, physics, and state transitions.
    • Frame-rate analysis checks for frame-rate independence across 30, 60, and 144 FPS simulations.
    • Deterministic simulation attempts to reproduce bugs using fixed seeds and scripted inputs.
    • Structured reporting produces severity-ranked bug reports with exact reproduction steps and suggested fixes.

    How it works

    1. Rule Analysis: The agent reads existing game rules or infers them from the source to establish a testing baseline.
    2. Code Mapping: It identifies critical paths including the main loop, state machines, and collision logic.
    3. Execution: The agent runs logic tests and observes behavior for frame-rate issues, input snagging, or memory growth.
    4. Reporting: Findings are documented in a BUG_REPORT.md categorized by severity and impact.

    Frameworks & tools

    Primary focus on web-based game engines, specifically Three.js and vanilla JavaScript/TypeScript game loops. Works with standard unit testing frameworks for logic validation.

    Why this beats prompting it yourself

    Generic prompts often miss "feel" bugs like corner snagging or frame-rate dependencies. This skill enforces a rigorous QA checklist covering input focus loss, entity cleanup, and determinism that manual prompting typically overlooks.

    Use cases

    • Physics validation to catch tunneling at high speeds or clipping through geometry.
    • State management to ensure score resets and entity cleanup work perfectly on restart.
    • Input stress-testing to find bugs caused by simultaneous key presses or window resizing.
    • Performance audits to monitor draw call growth or memory leaks during long sessions.

    Known limitations

    Requires project files to be accessible for code mapping. Playtesting observation is limited to available browser tools or console logs.

    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.

    Requires access to game source code. Cannot physically "feel" input latency or controller haptics. Deterministic testing requires game support for fixed seeds.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 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

    2 uses

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 25 days ago

    • Free with an Agensi account

    Permissions

    Read Files
    Write Files
    Terminal / Shell

    File Scopes

    src/**
    RULES.md
    BUG_REPORT.md

    Reads the game's source files, writes RULES.md and BUG_REPORT.md, and runs local tests. No network access needed.

    Listed25 days ago

    What's inside

    Frequently Asked Questions

    This skill is optimized for web-based games, specifically those built with Three.js, TypeScript, and vanilla JavaScript game loops, providing deep analysis of physics, state machines, and frame-rate dependencies.

    It identifies logic errors, collision math bugs like 'tunneling,' memory leaks, and frame-rate inconsistencies that standard AI prompts or manual playtesting often overlook.

    Simply provide the agent with access to your project's source code and any existing documentation (like a RULES.md); it will automatically map the game loop and critical logic paths.

    Yes, the agent generates a structured BUG_REPORT.md that ranks issues by severity and provides exact reproduction steps along with suggested code fixes.

    While it excels at testing logic, physics, and performance, it requires access to the underlying project files and cannot 'see' visual art assets or test purely visual aesthetics.

    New in Game Development

    Free