Skip to main content
Formal Specification

Gibraltar Hopscotch — Formal Specification

Gibraltar · dexterity-small · Confidence: low

← Back to game page

Implementation Suitability

Suitability: engine-readyFidelity: probabilistic-proxyGap-fills: 1
# Rayuela — AI-optimized specification ## General idea A sequential level-progression jumping game where agents advance through 8 discrete stages. On each stage `L in 1..8`, the active agent must land a marker in square `L`, navigate a hopscotch graph while skipping square `L`, retrieve the marker, and return to base without foul. ## Formal model - **Abstraction taken:** The 8-cell physical track is modeled as a directed topological graph with node types `Single(1..3, 6)`, `Pair(4-5, 7-8)`, and `Terminal(Cielo)`. The physical toss and single-leg balance checks are abstracted into discrete stochastic success probabilities parameterized by agent skill and level difficulty. - **Player/agent count:** 2 to 6 players. - **Turn model:** Strictly sequential turn rotation. ## Formal rules R1. Setup: Each player `i` starts with progression level `level[i] = 1`. R2. On player `i`'s turn: R2.1. Target square `T = level[i]`. R2.2. Toss check: The agent tosses the marker at square `T`. The resolver samples `P(toss_success | T, skill)`. R2.3. If toss fails (marker misses or touches line), turn ends immediately with no level advance. R2.4. Traversal check: If toss succeeds, the agent executes forward hops through nodes 1..8 (skipping `T`) to `Cielo`, turns around, and executes reverse hops. R2.5. At node adjacent to `T`, retrieval check: The resolver samples balance stability `P(retrieve_success | skill)`. R2.6. If any hopping step or retrieval fails, turn ends immediately with no level advance. R2.7. If the entire forward-return trajectory succeeds, `level[i]` increments by 1: `level[i] += 1`. R2.8. [gap-fill] Turn bonus: A successful level completion grants the active player an immediate consecutive attempt at the next level, up to a maximum of 2 consecutive successes per turn. R3. Win condition: The first player to successfully complete `level[i] == 8` wins the game immediately. ## Edge cases - Tie at the end condition: Not applicable — turns are strictly sequential; the first player to complete level 8 wins instantly. - A player with no legal move: Not applicable — tossing at the target square is always available. - Repetition / cycles: Bounded because `level[i]` is monotonically non-decreasing and cannot decrease upon fouls. - Simultaneous claims on the same resource or position: Not applicable — single active agent on the track at any time. - Player elimination mid-round: If a player disconnects, they are skipped in turn rotation; other players retain their progress. - Resource exhaustion: Not applicable — the marker is retrieved on every turn. - Termination guarantee: Guaranteed by monotonic level advancement `P(advance) > 0` on every turn attempt. ## State representation - Persistent state: `level` (array of ints 1..8 indexed by player_id), `current_player_idx` (int 0..N-1). - Per-turn ephemeral state: `toss_outcome` (boolean), `traversal_outcome` (boolean), `consecutive_wins_this_turn` (int 0..2). - Observable information: Fully observable by all players. ## Action space - `toss(target_square)`: Execute throw with calibration parameters. - `hop_step(node_id, foot_mode)`: Execute hop onto single node or split-pair node. - `retrieve_marker(target_square)`: Balance check to retrieve marker. ## Playability notes - Single-player-vs-AI: Natural — a solitaire race against an AI opponent tracking who finishes level 8 in fewer turn rotations. - Group games (n>4): Clean round-robin rotation with zero coalition complexity. - Physical-skill games: Motor balance and accuracy abstracted into parameterized success probabilities. - What is lost: The neuromuscular challenge of hopping on a single foot on uneven flagstones. ## Implementation hints Graph-based state machine is trivial to implement. Focus on tuning the error distribution `P(foul | level)` so higher numbered squares carry appropriately elevated risk.

Download as JSON

Download the full game corpus entry for this game as structured JSON.

Download JSON
Cite this page

Gibraltar Hopscotch (Gibraltar). (n.d.). Games of Countries. https://gamesofcountries.com/games/gi-gibraltar/rayuela-gibraltar

@misc{gamesofcountries-gi-gibraltar-rayuela-gibraltar,
  title = {Gibraltar Hopscotch (Gibraltar)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/gi-gibraltar/rayuela-gibraltar}
}