Counts show all available guides. Filters combine to narrow your results.
Simplify a complex taskGuide
Find the rule that can make your homemade game get stuck
Find a reachable loop before treating a homemade game as finished.
Illustrative resourceNot tool-testedAdults designing tiny original non-wagering games
What you’re working towards
A bounded state audit of a complete tiny non-wagering game, showing reachable stalls, endless cycles and minimally changed rule alternatives. This is a written practice result, not a recorded AI run.
Your choice also saves this guide’s current worked-example checkmarks. Only this browser; no account or sync. View saved guides
Saving needs JavaScript. Reading and downloading this guide does not save it.
Make a start
One step at a time.
3 steps
Start with the synthetic worked example or other non-sensitive inputs. Run one prompt at a time in your chosen AI tool. Replace any bracketed placeholders with the requested material, check each reply, and supply the source or earlier outputs that step needs. Paste them again if the tool lacks that context.
Start with the synthetic example, then adapt each prompt to your task.
01
Try the complete example
Open a text AI chat you already have permission to use. Paste the whole prompt below, from TASK through END INPUT, and send it. The input is source material; do not paste the illustrative Output as an answer to imitate. Expect A bounded state audit of a complete tiny non-wagering game, showing reachable stalls, endless cycles and minimally changed rule alternatives. If you receive advice instead of the result, ask for the complete requested result from the supplied input.
Your starting prompt
TASK
Enumerate every reachable state of this exact tiny game, including next player and last remover. Identify its terminal condition and any repeated-state cycle. Offer two clearly separate minimal rule repairs, state their exact wording and recheck both from the supplied start. Do not silently apply a repair to the original audit or claim game balance.
INPUT
Complete original non-wagering game. Two players A and B alternate; A starts. The pile begins with 3 tokens. On a turn remove exactly 2 tokens if at least 2 remain; otherwise pass. The game ends only when the pile reaches 0; the last remover wins. No other moves or endings exist. A state must include pile size, player whose turn is next and last remover (none initially). Examine all states reachable from this one start; do not claim a result for every possible game. Show a repeated-state counterexample if ending is not guaranteed. Propose two separate minimal rule repairs and recheck each from the same start. Termination does not establish fairness or fun.
END INPUT
02
Check and correct
Use three paper counters to trace 3 → 1 → 1 → 1. The pile/player/last-remover state (1,B,A) repeats. The original reachable set has exactly three states and no terminal state. Recheck each proposed repair separately from pile 3 with A starting: follow that repair’s exact rule wording and report the actual terminal state, turn count and winner. The repair labels and outcomes need not match the illustrative alternatives. Keep changes out of the original audit and do not treat termination as fairness or playability; a real playability trial remains needed. Compare with the unchanged source yourself. If anything differs, use the correction prompt, then recheck the complete replacement before saving.
Your starting prompt
Check the answer in this sample chat against the complete INPUT above. Use three paper counters to trace 3 → 1 → 1 → 1. The pile/player/last-remover state (1,B,A) repeats. The original reachable set has exactly three states and no terminal state. Recheck each proposed repair separately from pile 3 with A starting: follow that repair’s exact rule wording and report the actual terminal state, turn count and winner. The repair labels and outcomes need not match the illustrative alternatives. Keep changes out of the original audit and do not treat termination as fairness or playability; a real playability trial remains needed. Quote each mismatch and its source, then return a complete corrected replacement. If source or answer is missing, ask for it. Do not describe a suggested human trial as completed.
03
Use your own source instead
Open a new chat and copy the prompt below. Replace every bracketed field with your own permitted non-sensitive material, or write unknown. Change the source names, counts, dates and rules together wherever they occur. Expect a complete result using those new values. Compare each source link and calculation with your new input, using the review instructions in this prompt. If a fact or rule is wrong, quote the mismatch and source and ask for a complete corrected replacement; keep essential unknowns held rather than borrowing the practice values.
Your starting prompt
Audit my complete tiny non-wagering game from its stated initial condition. Define a state using every value that affects legal moves, turn order or winner. Enumerate the bounded reachable states and legal transitions; mark actual terminal states. Show an exact repeated-state path for a cycle or an exact dead end for a stall. State the scope and do not imply an exhaustive audit if the state space is too large or rules are incomplete. Propose separate minimal repairs with exact wording and recheck each under my original starting condition. Review moves against the source rules and keep termination separate from fairness or enjoyment.
MY INPUT
Complete original game rules: [replace with your source or rule]
Initial state: [replace with your source or rule]
Legal actions and turn rules: [replace with your source or rule]
Terminal condition: [replace with your source or rule]
Small state-space boundary: [replace with your source or rule]
Use only MY INPUT for facts, quantities, names, dates and permissions. Ask for missing essential facts or conflicting rules instead of guessing. Creative invention is allowed only where this task and my boundaries permit it. Return the actual complete requested result, a separate unknown/hold list and the source-specific review described above. Do not reuse practice-example facts. Keep this a draft; do not publish, contact people, book, buy or change another service.
Optional: improve the result
Check a changed starting pile
Optional, sample-only extension. Return to the practice chat, or paste this guide’s complete practice input and your checked answer into a new chat. Test the boundary of the original finding. This does not apply unchanged to your own source.
Advanced prompt
For the original sample rules only, separately trace starting piles 2 and 4, with A starting and no last remover. Give every transition, terminal condition and winner. Explain why those cases do not repair the original pile-3 cycle. If the complete sample input or answer is missing, ask me to paste it. Keep new variants separate from the original result.
Show, then try
See the shape of the result.
This input is synthetic. The output is a written illustration of the intended result, not a verified AI-tool run.
01 / The starting point
Synthetic input
COMPLETE SYNTHETIC PRACTICE INPUT
Complete original non-wagering game. Two players A and B alternate; A starts. The pile begins with 3 tokens. On a turn remove exactly 2 tokens if at least 2 remain; otherwise pass. The game ends only when the pile reaches 0; the last remover wins. No other moves or endings exist. A state must include pile size, player whose turn is next and last remover (none initially). Examine all states reachable from this one start; do not claim a result for every possible game. Show a repeated-state counterexample if ending is not guaranteed. Propose two separate minimal rule repairs and recheck each from the same start. Termination does not establish fairness or fun.
Copies text only. For image or file tasks, use the source files named in the steps too.
ILLUSTRATIVE AUTHORED OUTPUT
Written for this guide; no consumer AI execution or real-world trial is claimed.
ORIGINAL RULE AUDIT
Terminal condition: pile = 0, with the last remover winning.
S0: (pile 3, next A, last none).
A removes 2 → S1: (pile 1, next B, last A).
B must pass → S2: (pile 1, next A, last A).
A must pass → S1: (pile 1, next B, last A), exactly the earlier state.
Reachable distinct states: S0, S1, S2. None has pile 0. Because each state has only one legal move, this is the complete reachable state graph for the supplied start. The S1 → S2 → S1 cycle continues forever. The original game does not end from this start.
SEPARATE REPAIR A — CHANGE THE MOVE RULE
Replace the removal rule with: “Remove 2 tokens when at least 2 remain; remove 1 when exactly 1 remains.” Keep the original pile-zero ending and last-remover winner.
Recheck: (3,A,none) → A removes 2 → (1,B,A) → B removes 1 → terminal pile 0, winner B. Two removal turns; no passes.
SEPARATE REPAIR B — CHANGE THE ENDING RULE
Keep the original move rule. Add: “After a removal, if fewer than 2 tokens remain, end immediately; the last remover wins.” Keep the original pile-zero ending for any case that reaches it.
Recheck: (3,A,none) → A removes 2 → terminal pile 1, winner A. One turn. B never takes a turn under this repair.
These are alternatives, not simultaneous amendments. Repair B changes the winner and can be unsatisfying despite ending. Neither repair is a balance or enjoyment claim. A human paper-token trial remains unperformed.
These checks use the worked example and its particular facts. With your own material, check your actual names, numbers, sources and rules; do not reuse the sample’s values as the answer.
Unsaved checks stay in this tab and reset when you reload. If you choose Save guide, its current example checkmarks are kept in this browser when storage is available.
!
A few things to watch for.
Checking only pile size can miss whose turn it is. Include all state fields that determine legal actions or winners.
A rule change that ends the game may also change who wins. Present that consequence, not a claim of improved balance.