Token导航 LogoToken导航TokenDH.com
前端设计需要联网github未标认证来源可访问clear审计通过

adding-support-mod-parsers添加支持 mod 解析器

Agent Skill

用于辅助前端页面、组件、样式和交互逻辑的开发与维护。它适合让 Agent 生成或审查 React、Next.js、Vue、Tailwind、CSS 等相关代码,整理组件结构,或定位布局和性能问题。使用时需要结合项目现有设计系统、路由和构建方式,避免只生成孤立片段;涉及页面改动时,应配合本地预览和构建检查确认视觉效果。

总安装

649

周安装

26

GitHub Stars

26

下载量

210
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:adding-support-mod-parsers(添加支持 mod 解析器)
来源仓库:https://github.com/aclinia/torchlight-of-building
仓库路径:skills/adding-support-mod-parsers
安装命令:
npx skills add https://github.com/aclinia/torchlight-of-building --skill adding-support-mod-parsers
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/aclinia/torchlight-of-building --skill adding-support-mod-parsers

简介

用于解析支持技能(support skill)的修饰符字符串并转换为结构化对象。

  • 适用于游戏或技能系统中处理被动/主动技能的等级缩放逻辑。
  • 可支持新增支持技能修饰符模式或扩展现有解析规则。
  • 需结合项目中的技能模板和类型定义进行集成,注意运行时性能影响。
  • adding-support-mod-parsers 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Adding Support Mod Parsers

Overview

Support mod parsers convert raw support skill affix strings (e.g., "+15% additional damage for the supported skill") into typed SupportMod objects at runtime. Unlike active/passive skills which use level-scaling factories, support skills parse their affixes directly using templates.

When to Use

  • Adding support for new support skill affix patterns
  • Extending support mod parsing to handle new variants

Project File Locations

PurposeFile Path
Support mod templatessrc/tli/skills/support-mod-templates.ts
Mod type definitionssrc/tli/mod.ts
SupportMod typesrc/tli/core.ts
Template/spec helperssrc/tli/mod-parser/
Calculation handlerssrc/tli/calcs/offense.ts

Implementation Checklist

1. Check if Mod Type Exists

Look in src/tli/mod.ts under ModDefinitions. If the mod type doesn't exist, add it first (see adding-mod-parsers skill).

2. Add Template in support-mod-templates.ts

Templates use the same DSL as the main mod parser:

// In allSupportParsers array
t("{value:dec%} additional damage for the supported skill").output(
  (c) => ({
    type: "DmgPct",
    value: c.value,
    dmgModType: "global",
    addn: true,
  }),
),

Template capture types:

TypeMatchesExample Input → Output
{name:int}Unsigned integer"5"5
{name:dec}Unsigned decimal"21.5"21.5
{name:int%}Unsigned integer percent"30%"30
{name:dec%}Unsigned decimal percent"96%"96
{name:+int}Signed integer (requires + or -)"+5"5, "-3"-3
{name:+dec}Signed decimal (requires + or -)"+21.5"21.5
{name:+int%}Signed integer percent"+30%"30, "-15%"-15
{name:+dec%}Signed decimal percent"+96%"96

Signed vs Unsigned Types:

  • Use unsigned (dec%, int) when input does NOT start with + or - (e.g., "0.8% additional damage")
  • Use signed (+dec%, +int) when input STARTS with + or - (e.g., "+19.8% additional damage")
  • Signed types will NOT match unsigned inputs, and vice versa
  • IMPORTANT: Some support skills have signed inputs, others have unsigned - you may need BOTH templates (see examples below)

Optional syntax:

  • [additional] - Optional literal, sets c.additional?: true
  • (effect|damage) - Alternation (regex-style)
  • \\( and \\) - Escaped parentheses for literal matching

3. SupportMod Structure

Each parsed mod is wrapped in SupportMod:

interface SupportMod {
  mod: Mod;
}

The parseSupportAffix function handles this wrapping:

return mods.map((mod) => ({ mod }));

4. No-Op Parsers (Informational Text)

Use outputNone() when a mod string should be recognized (not flagged as unparsed) but has no effect on calculations:

t("always attempts to trigger the supported skill. interval: {_:dec}s").outputNone(),
t("automatically use the supported attack skill to continuously attack the nearest enemy within {_:int}m while standing still").outputNone(),

IMPORTANT: Use outputNone(), NOT outputMany([]). Both work but outputNone() is the correct API for this purpose.

5. Multi-Output Parsers

For affixes that produce multiple mods:

t("{value:dec%} additional attack and cast speed for the supported skill")
  .outputMany([
    spec((c) => ({ type: "AspdPct", value: c.value, addn: true })),
    spec((c) => ({ type: "CspdPct", value: c.value, addn: true })),
  ]),

6. Add a Test

Add a test case to src/tli/skills/support-mod-templates.test.ts using the example input string given to you:

test("parse <skill name> <description of what it parses>", () => {
  const result = parseSupportAffixes([
    "<exact input string from the skill data>",
  ]);
  expect(result).toEqual([
    [
      {
        mod: {
          type: "<ModType>",
          // ... expected mod properties
        },
      },
    ],
  ]);
});

7. Verify

Run tests to ensure parsing works:

pnpm test
pnpm typecheck
pnpm check

Examples

Damage Mod with BOTH Signed and Unsigned Variants

Some support skills use +{value}% templates (e.g., Increased Area) while others use {value}% (e.g., Haunt). You need BOTH templates:

Inputs:

  • "+19.8% additional damage for the supported skill" (Increased Area - signed)
  • "0.8% additional damage for the supported skill" (Haunt - unsigned)
// Signed version (e.g., "+19.8% additional damage...")
t("{value:+dec%} additional damage for the supported skill").output(
  (c) => ({ type: "DmgPct", value: c.value, dmgModType: "global", addn: true }),
),
// Unsigned version (e.g., "0.8% additional damage...")
t("{value:dec%} additional damage for the supported skill").output(
  (c) => ({ type: "DmgPct", value: c.value, dmgModType: "global", addn: true }),
),

Typed Damage Mod (Signed)

Input: "+20% additional melee damage for the supported skill"

t("{value:+dec%} additional melee damage for the supported skill").output(
  (c) => ({ type: "DmgPct", value: c.value, dmgModType: "melee", addn: true }),
),

Attack Speed (Signed - can be negative)

Input: "-15% Attack Speed for the supported skill" (Steamroll)

t("{value:+dec%} attack speed for the supported skill").output(
  (c) => ({ type: "AspdPct", value: c.value, addn: false }),
),

Note: +dec% matches both + and - signs.

Conditional Mod

Input: "The supported skill deals +30% additional damage to cursed enemies"

t("the supported skill deals {value:dec%} additional damage to cursed enemies")
  .output((c) => ({
    type: "DmgPct",
    value: c.value,
    dmgModType: "global",
    addn: true,
    cond: "enemy_is_cursed",
  })),

Per-Stackable Mod

Input: "+5% additional damage for the supported skill for every stack of buffs while standing still"

t("{value:dec%} additional damage for the supported skill for every stack of buffs while standing still")
  .output((c) => ({
    type: "DmgPct",
    value: c.value,
    dmgModType: "global",
    addn: false,
    per: { stackable: "willpower" },
  })),

Mod with No Value

Input: "The supported skill cannot inflict wilt"

t("the supported skill cannot inflict wilt").output(() => ({ type: "CannotInflictWilt" })),

Escaped Parentheses

Input: "Stacks up to 5 time(s)"

t("stacks up to {value:int} time(s)").output((c) => ({
  type: "MaxWillpowerStacks",
  value: c.value,
})),

Note: Literal ( and ) don't need escaping when they don't contain alternations.

Complex Pattern with Ignored Values

Input: "When the supported skill deals damage over time, it inflicts 10 affliction on the enemy. Effect cooldown: 3 s"

t("when the supported skill deals damage over time, it inflicts {value:int} affliction on the enemy. effect cooldown: {_:int} s")
  .output((c) => ({ type: "AfflictionInflictedPerSec", value: c.value })),

Use {_:type} to capture but ignore values.

Shadow Quantity (Signed Flat Integer)

Input: "+2 Shadow Quantity for the supported skill"

t("{value:+int} shadow quantity for the supported skill").output(
  (c) => ({ type: "ShadowQuant", value: c.value }),
),

Template Ordering

IMPORTANT: More specific patterns must come before generic ones in allSupportParsers array.

// Good: specific before generic
t("{value:dec%} additional melee damage for the supported skill").output(...),
t("{value:dec%} additional damage for the supported skill").output(...),

// Bad: generic would match first
t("{value:dec%} additional damage for the supported skill").output(...),
t("{value:dec%} additional melee damage for the supported skill").output(...),  // never matches

Common Mistakes

MistakeFix
Using dec% for input with + prefixUse +dec% for inputs like "+25% damage"
Using +dec% for input without signUse dec% for inputs like "0.8% damage"
Only one template when inputs varyAdd BOTH signed and unsigned templates (see examples)
Generic template before specificMove specific templates earlier in array
Missing type field in output mapperInclude type: "ModType" in the returned object
Handler doesn't account for new mod typeUpdate offense.ts to handle new mod types
Forgot the wrapper structureparseSupportAffix already wraps in {mod}
Using outputMany([]) for no-opUse outputNone() instead

Data Flow

Support skill affix: "+15% additional damage for the supported skill"
    ↓ parseSupportAffixes()
    ↓ normalize (lowercase, trim)
"15% additional damage for the supported skill"
    ↓ template matching (allSupportParsers)
[{ mod: { type: "DmgPct", value: 15, dmgModType: "global", addn: true } }]
    ↓ resolveSelectedSkillSupportMods() in offense.ts
Applied to skill calculations

Difference from Main Mod Parser

AspectMain Mod ParserSupport Mod Parser
Filesrc/tli/mod-parser/templates.tssrc/tli/skills/support-mod-templates.ts
SourceGear affixes, talents, etc.Support skill affixes only
OutputMod[]SupportMod[] (wrapped in {mod})
UsageparseMod()parseSupportAffixes()

Both use the same template DSL (t(), spec(), outputMany(), outputNone()).

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.13%
按下载量换算70

replit

21.73%
按下载量换算46

windsurf

19.54%
按下载量换算41

OpenCode

11.73%
按下载量换算25

Cursor

7.71%
按下载量换算16

Claude Code

4.01%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills