Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    Game Designer Specialist

    1

    Turn your game idea into exact mechanics, meaningful player choices and a clear rules packet your developer can build.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Game Designer Specialist

    Game Designer Specialist

    Example session with this skill installed

    Design a turn-based blink mechanic for an original offline grid puzzle. Confirmed rules: the player has a capacity of three charges; a legal blink costs one charge and moves exactly three clear tiles in a cardinal direction; the destination and every crossed tile must be clear. A blink starts a cooldown of two subsequent accepted gameplay commands. Invalid commands do not advance a turn or cooldown, and pause freezes all gameplay state. Normal movement is an accepted command. Charges do not regenerate. We have no engine, prototype or playtest. Produce the actual rule packet, denial behavior, contrasting traces and criteria. Interruption, reset and UI presentation are not settled; keep those as explicit proposals. Do not add combat or a metagame.

    • Read your context and instructions
    • Compiled the game designer specialist

    Illustrative preview: this representative editorial result is not a recorded runtime session.

    Make blink a deliberate route choice with an exact, inspectable transition. Keep charge and cooldown semantics separate: an unavailable blink must not spend a charge or advance a turn.

    Mechanics packet BLINK-01 / version 1
    Producer: Game Designer Specialist. Recipient: implementation owner.
    Status: COMPLETE_LOCAL for the rule artifact; interruption/reset decisions remain NEEDS_DECISION. No engine or player check was executed.
    Objective: specify the supplied movement mechanic. Scope: offline turn-based grid rules; no combat, charge regeneration or metagame.
    Inputs: supplied capacity, cost, distance, clear-path, cooldown, invalid-command and pause rules. Proposed IDs below are local.

    RulePrecondition / eventState transitionDenial / feedback
    BL-R1 Blinkactive gameplay; charges >= 1; cooldown = 0; cardinal direction; all three crossed/destination tiles clearmove three tiles; charges -= 1; cooldown = 2; accept exactly one gameplay commandA failed prerequisite changes no position, charge, turn or cooldown; communicate the actual reason
    BL-R2 Ordinary moveaccepted under the supplied game's normal-movement rulesapply that movement; if cooldown > 0, decrement it onceDo not invent normal-movement collision semantics beyond the supplied contract
    BL-R3 Invalid actionany denied commandall gameplay state unchangedA denial may show a message; it never counts as a recovery turn
    BL-R4 Pausepaused modereject gameplay progression; retain position/charges/cooldownPresent the paused state; resume must not silently replay a rejected input

    Ordering: BL-R1's accepted blink sets cooldown to two; that same command does not decrement its newly set cooldown. The next two accepted gameplay commands decrement 2→1→0. Only then can another blink be accepted. Game state, rather than a HUD animation, owns availability.

    Contrasting traces under these rules
    A: (charges 3, cooldown 0, clear path) → Blink → (2,2); ordinary move → (2,1); ordinary move → (2,0); clear-path Blink → (1,2).
    B: (2,2) → Blink denied by cooldown → (2,2); invalid ordinary move → (2,2); accepted ordinary move → (2,1). Neither denial advances recovery.
    C: (1,0, second crossed tile blocked) → Blink denied → (1,0), same position. A clear destination alone is insufficient.
    D: (0,0) → Blink denied by empty charges → (0,0). No regeneration is implied.
    E: pause at (2,1) → attempted move/blink → unchanged; resume → (2,1) before the next newly accepted command.
    These are specification traces, not recorded program output.

    Proposals requiring an owner decision

    • Atomicity: validate and commit one blink as a single transition; animations present that result. An interrupted animation should not partially spend or refund a charge unless a different explicit interruption rule is accepted.
    • Reset: a fresh level starts at supplied initial state with three charges and zero cooldown. Resume/load preserves saved state. This reset policy is proposed, not confirmed.
    • Presentation: label charge count and remaining accepted commands; preview the three-tile path with blocked-path feedback. Exact visual language is unchosen.

    Acceptance mapping
    BL-AC1/R1: clear legal blink changes position exactly three tiles, spends exactly one charge and sets cooldown to two.
    BL-AC2/R1,R3: blocked intermediate tile, wrong direction, empty charge and cooldown denial preserve all gameplay state.
    BL-AC3/R2: only the next two accepted commands reduce cooldown; invalid actions do not.
    BL-AC4/R4: pause preserves state, with no replayed gameplay action on resume.
    BL-AC5: confirm interruption/reset policy before dependent integration; test save/reload under that policy.

    Experience hypothesis: limited blinks create a route-planning trade-off instead of a universally superior move. Smallest useful experiment: original grid rooms with a viable walking route and more than one occasion to spend charges. Record chosen route, charge timing, explanation of cooldown and softlock/recovery behavior on a versioned prototype. An all-blink tactic or consistently unused blink can challenge the hypothesis. No participant result, balanced value or enjoyment claim is supplied.
    Next action: owner confirms the two material policies; developer implements the packet and binds functional proof to BL-AC1–5. Human observation answers comprehension and choice separately.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Make the next player choice worth designing around.

    Game Designer Specialist turns a promising concept into the rules that make it playable. Use it to develop a core loop, repair a mechanic, shape progression or examine a dominant strategy before it spreads through the rest of your game.

    The method connects each player action to its prerequisites, cost, state change, timing, feedback and recovery. It works through contrasting situations so the specification answers what happens when a resource runs out, an action is interrupted or two events compete. Your developer receives concrete behavior instead of another paragraph about how the mechanic should feel.

    You can get

    • A filled mechanics packet with legal actions, denial behavior and worked examples.
    • Progression and resource rules with explicit incentives and tuning options.
    • A focused experiment that tells you what to observe and which design choice the result should change.

    For example, bring a limited-use movement ability. The specialist can define charge spending, cooldown boundaries and failed attempts, then examine whether the ability creates a real route choice. Proposed numbers stay visible as design choices you can adjust.

    Start with your intended player experience, genre, confirmed rules and constraints. The complete method and reference work independently of the other Game Squad roles. You supply AI access; a specification can be produced without an engine, while playable prototypes and balance decisions require the appropriate tools and player observations.

    How to use

    Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed 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 scanned

    Verified clean 16 days ago

    • Passed all security checks, Safe to install

    Listed16 days ago

    What's inside

    Frequently Asked Questions