| name | orryx-progression-curve-skill |
| description | Design and simulate Orryx experience, mana, spirit, and upgrade-point progression. Use for ExperienceOfLevel curves, job resource formulas, level-by-level balance tables, monotonicity and boundary diagnostics, and deterministic line-chart artifacts based on experiences/default.yml and jobs/剑修.yml. |
| license | MIT |
| activation | {"command":"/orryx-progression-curve-skill","intents":["design Orryx experience curve","simulate Orryx job progression","chart mana spirit or upgrade points"]} |
| metadata | {"author":"NarraFork","version":"1.1.0","created":"2025-07-13T00:00:00.000Z","last_reviewed":"2025-07-13T00:00:00.000Z","review_interval_days":90} |
| provenance | {"repository":"Orryx","evidence":"references/source-evidence.md","contract":"references/contract.md"} |
/orryx-progression-curve-skill
生成经验曲线、职业资源公式、逐级模拟表和可复现折线图协议 artifact。
Orryx 字段
经验文件:
Options.Min
Options.Max,必须大于 Min
Options.ExperienceOfLevel,在 level 上下文中计算当前级所需经验
职业资源:
MaxManaActions / RegainManaActions
MaxSpiritActions / RegainSpiritActions
UpgradePointActions
Experience,引用经验文件 basename ID
真实默认经验曲线是 calc "200*level",范围 0..20。剑修 示例使用 MaxManaActions: calc "100+10*level"、固定精力上限 100、回复 5、每级升级点 1,并引用 default。
逐级模拟
对 Min..Max 每一级计算:
- 本级经验需求;
- 到达该级前的累计经验;
- 法力上限与回复;
- 精力上限与回复;
- 本级与累计升级点。
报告负值、非整数经验、断崖式跳变、意外下降、零值区间和溢出风险。边界级必须纳入模拟。
line-chart artifact protocol
折线图不是二进制图片,而是确定性 JSON artifact:
{
"kind": "line-chart",
"title": "Progression by level",
"x": {"name": "level", "values": [0, 1, 2]},
"series": [{"name": "experience", "values": [0, 200, 400]}],
"units": {"experience": "xp"}
}
要求:level 升序、每个 series 与 x 等长、数值有限、名称稳定、无生成时间。artifact 的 mediaType 为 application/vnd.orryx.line-chart+json。
工作流
- 读取范围、公式、资源规则和目标节奏。
- 逐级求值并累计,不只抽样首尾等级。
- 生成经验 YAML、可选职业片段、模拟表与折线图 artifacts。
- 将经验引用与异常曲线写入 diagnostics/checks。
- 不调用游戏服务器或执行 reload。
运行
py -3 scripts/run_pipeline.py --input assets/request.example.json --output result.json