Skip to content
Back to skills

Balance Design

ASecurity

Use when comparing options, matchups, roles, economies or catch-up systems and the project needs explicit balance goals, metrics, tradeoffs and versioned evidence instead of universal fairness rules.

  • 11 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 23, 2026
ai-agentsgorails

Security analysis

A100/100

Scanned October 23, 2026

npx -y skills add jammyfu/open-game-skills --skill balance-design --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Balance Design?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Balance Design
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jammyfu-balance-design/badge)](https://www.skillsdirectory.com/skills/jammyfu-balance-design)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: balance-design
description: Use when comparing options, matchups, roles, economies or catch-up systems and the project needs explicit balance goals, metrics, tradeoffs and versioned evidence instead of universal fairness rules.
---

# Balance design

Choose the project goal before changing numbers.

## Contract

Publish:
- stable `balance_policy_id` and revision
- population/mode/build scope
- primary metric definition and denominator
- secondary guardrails
- compared option or matchup IDs
- authored tradeoffs and allowed asymmetry
- evidence window and confidence/coverage limits

Balance may target viability, role separation, pacing, accessibility, competitive parity, progression, spectacle, or another stated project goal. No one target is universal.

## Ownership

This skill owns balance policy and evidence interpretation. It consumes combat/economy/race/system values from their true owners; it does not silently rewrite hitstop, input timing, camera, collision, grants or other system contracts.

## Rules

1. Compare like-for-like contexts; a global average without mode/rank/loadout denominator is not evidence.
2. Distinguish intentional asymmetry from accidental dominance.
3. Counter strength, comeback systems and catch-up behavior are project policies, not universal moral rules.
4. Every tuning change records before/after policy revision, changed values and expected metric movement.
5. If a change alters another owner's contract, route the change there and rerun its acceptance tests.

## Acceptance

A reviewer can identify the balance objective, denominator, compared IDs, intentional tradeoffs, changed revision and evidence required to keep or revert the change. A different design philosophy can use the same contract without inheriting hidden genre assumptions.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.