skilld
All skills
thedivergentai avatar

/godot-monte-carlo-balancer

@9d6e91e
by Divergent AIthedivergentai/gd-agentic-skills783 stars
48

Use when auditing or recalibrating game balance: build a source-driven Monte Carlo balance lab (Rust + rayon) that extracts live game data, simulates human playstyles (AFK→pro), emits win-rate/economy verdicts with confidence intervals, and bruteforce-tunes parameters. Trigger on unfair levels, unreachable shops, farm exploits, interest-curve cliffs, post-content recalibration, or CI balance JSON diffs. Keywords: balance lab, Monte Carlo, win rate, difficulty curve, economy career, playstyle simulation, Resource extraction, GDScript parser, bruteforce tuning.

Ask your Agent

Use this Skill: https://skilld.dev/gh/thedivergentai/gd-agentic-skills/godot-monte-carlo-balancer

This session only. Nothing lands on disk.

referencesexample-lane-defense.md

≈749 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Reference specialization — Lane Defense

Status: Example only. Do not use these field names as the default GameData / RunResult shape. Map them from the abstract model in Phase 0 when the target game is lane defense / shift-based tower-shooter.

Contributed patterns originated in a Godot 4.x lane-defense Balance Lab (weapons, traps, jams/overheat, barricade HP, wave leaks).

Abstract → lane-defense map

Abstract Lane defense
Session One shift (wave sequence)
Threat Enemy waves vs barricade
Defense Weapon + traps + upgrades
Faults Jams / overheat minigames
Resources Ammo / supplies / heat
In-run economy Cash → traps/upgrades
Meta economy Stars → store
Grade Stars by HP remaining

Example PlayStyle extensions

Lane defense adds trap/upgrade spend axes on top of core reaction params:

  • places_traps, trap_budget_fraction
  • buys_upgrades, upgrade_cash_reserve
  • Fault reaction vs supply reaction as separate mean/sigma pairs
  • Reference afk→pro table (tune per game): slower reactions / higher lapse for afk; higher trap budget for pro

Example RunResult texture fields

barricade_hp, leaks_by_kind, jams, overheats, fault_downtime, empty_downtime, ammo_wasted, supplies_missed, traps_placed, coins_new_best vs coins_replay, peak_heat_ratio, time_to_first_fault.

Example sim update slices

update_spawning, update_supplies, update_cooling_and_clearing, update_firing, update_projectiles_and_traps, update_enemies, maybe_buy_upgrade, trap placement helpers.

Extraction smells from that reference

  • Factory create_*() -> WeaponProfile blocks in .gd (migrate to Resources when possible).
  • Inline HP scale: 1.0 + (a + b * level_id).
  • Mode rules inside match Kind.X: with rules.field = value.
  • Shop catalogs and trap-slot unlock arrays in progression scripts.

Example extract→inspect smell (lane defense)

After inspect: HP scale coeffs (0.08, 0.03) match wave_spawner.gd; weapon stapler.damage=12 matches factory; one enemy shows speed=1.0 (default!) → stop — regex missed the field; fix extract before any matrix. Do not invent a full sim from this file — use Phase 2 abstract types.

<!-- GDSkills research links (agents) Official docs: - https://docs.godotengine.org/en/stable/tutorials/scripting/resources.html - https://docs.godotengine.org/en/stable/classes/class_json.html - https://docs.godotengine.org/en/stable/tutorials/editor/command_line_tutorial.html Related skills: - https://github.com/thedivergentai/gd-agentic-skills/blob/main/skills/godot-resource-data-patterns/SKILL.md — Resource-first extract - https://github.com/thedivergentai/gd-agentic-skills/blob/main/skills/godot-testing-patterns/SKILL.md — Phase 7 headless calibration Parent skill: https://github.com/thedivergentai/gd-agentic-skills/blob/main/skills/godot-monte-carlo-balancer/SKILL.md -->

Source: SKILL.md on GitHub

Third-party checks

No alerts1mo3 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    This skill provides a robust framework and Rust-based templates for creating Monte Carlo game balance simulations in Godot 4.7. It includes automated build scripts, source-driven data extraction patterns, and statistical analysis tools. Security analysis confirms the skill follows safe practices, utilizing trusted external resources and standard development workflows.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

Provenance

Signed by skilld at 9d6e91e. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 3 weeks ago.

Activeupdated 2 months ago

README badge

README badge for thedivergentai/gd-agentic-skills/godot-monte-carlo-balancer