Design, inspect, or repair game mechanics, core loops, rules, goals, actions, state, chance, secrets, skills, and emergent dynamics.
SkillScore45Maintained
Automatic
◆Product Management pack
Last verified 8/16/2026
What it changes
Passed 10 of 22 test cases with the skill, and 3 of 22 without it.
Without skill14%
With skill45%
Checks it passed that the model alone missed
The Test Scenarios section provides deterministic test cases.
The Implementation Notes section explicitly includes instrumentation.
The Implementation Notes section details tunable constants and edge cases.
and 19 more in the full report
8 fixed2 passed either way11 still failing1 broke
Pass-rate lift
+32 pp
Tokens
+19%
Agent turns
0%
Response time
+10%
Model gemini-3.6-flashMeasured 8/16/2026v110 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
I'm designing an active-reloading mechanic for a sci-fi shooter and need a complete spec for our AI developer. Please write a document that includes the overall mechanic brief (core loop and player promise), a full rules model governing state and actions, technical implementation notes with data structures and tunable variables, potential failure modes such as degenerate loops, and deterministic test cases to verify the implementation.
What was checked, without and with the skill
Checked
Without skill
With skill
The response includes a top-level section titled Mechanic Brief containing player promise, core loop, and intended dynamics.
Missed
Passed
The response includes top-level sections titled Rules Model, Implementation Notes, Failure Modes, and Test Scenarios.
Missed
Passed
The Implementation Notes section explicitly includes instrumentation along with data structures and tunable constants.
Missed
Passed
case-01 · gemini-3.6-flash · 8/16/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md32 linessingle file
Game Mechanics Design
Use this skill to make gameplay systems explicit enough for an AI coding agent to implement, test, and tune. The mechanic spec should separate rule structure from presentation so the same idea can be prototyped cheaply before polish.
Source Traceability
Primary source: The Art of Game Design: A Book of Lenses, Third Edition by Jesse Schell, especially chapters 12-14 on mechanics, balance, and puzzles. The workflow is transformed and paraphrased.
Supporting source: MDA, which helps distinguish coded mechanics from the player-visible dynamics and aesthetic experience they produce.
Workflow
Identify the core loop: player input, system response, feedback, reward or consequence, and next decision.
Specify mechanics across space, time, objects, attributes, state, actions, rules, goals, skill, chance, and information visibility.
Predict dynamics: dominant strategies, degenerate loops, emergent interactions, pacing, and failure states.
Define player mastery: what starts simple, what becomes expressive, and how players improve.
Produce implementation-ready rules with test cases and tuning parameters.
Required Output
Mechanic Brief: player promise, core loop, and intended dynamics.
Rules Model: state, objects, actions, rules, goals, and information.
Implementation Notes: data structures, tunable constants, edge cases, and instrumentation.
Failure Modes: exploits, dead ends, unreadable outcomes, and boring optimal play.
Test Scenarios: deterministic cases the agent can implement.