Agent skill

Aaalice Refactor Parity Loop

by Aaalice233 in Aaalice233/Aaalice_NAI_Launcher

对 Aaalice NAI Launcher 的大型文件拆分型重构执行多子代理等价性闭环。当任务目标是把大型业务文件按职责拆分、同时保持功能、交互与布局不变时自动使用;不用于普通重构、功能开发、单点 Bug 修复或常规代码审查。

MITAuto-check passedDevelopment

Install Aaalice Refactor Parity Loop

skills CLI
$ npx skills add Aaalice233/Aaalice_NAI_Launcher --skill aaalice-refactor-parity-loop -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Aaalice233/Aaalice_NAI_Launcher aaalice-refactor-parity-loop --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Aaalice233/Aaalice_NAI_Launcher.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/aaalice-refactor-parity-loop .claude/skills/aaalice-refactor-parity-loop && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
aaalice-refactor-parity-loop
GitHub stars
188
Token cost
~1.3k tokens
SKILL.md length
282 words
Files
3 (incl. references)
Skills in repo
5
Repo updated
First seen
Licence
MIT

At a glance

对 Aaalice NAI Launcher 的大型文件拆分型重构执行多子代理等价性闭环。当任务目标是把大型业务文件按职责拆分、同时保持功能、交互与布局不变时自动使用;不用于普通重构、功能开发、单点 Bug 修复或常规代码审查。

  • Works in 2 steps: 当前任务是大型业务文件的职责拆分或同等规模的拆分型重构; → 目标要求重构前后功能、交互和布局保持等价,而不是引入新行为。
  • Tasks that involve Refactoring
  • SKILL.md covers 启动门禁, 适用边界, 强制原则 and 进度记录, plus 7 more sections
  • Calls git

What it does

Aaalice Refactor Parity Loop is an agent skill from Aaalice233/Aaalice_NAI_Launcher. 对 Aaalice NAI Launcher 的大型文件拆分型重构执行多子代理等价性闭环。当任务目标是把大型业务文件按职责拆分、同时保持功能、交互与布局不变时自动使用;不用于普通重构、功能开发、单点 Bug 修复或常规代码审查。

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/agent-prompts.md` and `references/progress-template.md`).

It sits in Development, covering Refactoring. It works with Git. The licence is MIT.

When your agent uses it

  • Tasks that involve Refactoring

Example prompts

  • “/aaalice-refactor-parity-loop”

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. 当前任务是大型业务文件的职责拆分或同等规模的拆分型重构;
  2. 目标要求重构前后功能、交互和布局保持等价,而不是引入新行为。

What it can do on your machine

Read from SKILL.md and the folder at commit fe0f12b. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Aaalice Refactor Parity Loop loads about 1.3k tokens when it runs, and up to ~2.9k if it reads all its reference files. Until then it costs about 36 tokens; SKILL.md has 282 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~36
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Aaalice233/Aaalice_NAI_Launcher at commit fe0f12b, republished under its MIT licence (© Aaalice233). 282 words, ~1,296 tokens.

Download SKILL.mdSave it as .claude/skills/aaalice-refactor-parity-loop/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
aaalice-refactor-parity-loop
description
对 Aaalice NAI Launcher 的大型文件拆分型重构执行多子代理等价性闭环。当任务目标是把大型业务文件按职责拆分、同时保持功能、交互与布局不变时自动使用;不用于普通重构、功能开发、单点 Bug 修复或常规代码审查。

Aaalice 重构等价性闭环

运行环境需要 Git、Flutter/Dart 与可用的 Codex 多代理协作工具。

本 Skill 验证的是“重构前后行为等价”,不是代码是否看起来更整洁。当前主 Agent 对范围、结论、修改、验证和最终交付负责;子代理只提供独立证据或执行边界清晰的修复。

本 Skill 可以自动启用,也可以由用户通过 $aaalice-refactor-parity-loop 明确调用。

启动门禁

自动启用时,只有同时满足以下条件才执行后续流程:

  1. 当前任务是大型业务文件的职责拆分或同等规模的拆分型重构;
  2. 目标要求重构前后功能、交互和布局保持等价,而不是引入新行为。

任一条件不满足时,不执行本 Skill 的多子代理循环,并简要说明它只用于大型文件拆分等价性验证。

先阅读:

适用边界

在以下场景使用:

  • 把超过 1000 行且经职责、内聚性和扩展成本评估后确有必要拆分的业务文件按职责拆分;
  • 移动 Widget、Controller、Provider、Service、路由或平台适配代码,但不应改变行为;
  • 重构后必须保持功能、状态、调用契约、响应式布局和平台行为;
  • 用户要求多子代理循环审查、对抗性确认和修复。

不在以下场景使用:

  • 有意改变产品行为的新功能;
  • 单点 Bug 修复、普通命名整理或机械格式化;
  • Changelog、Release Notes 或发布差异审查;
  • 仅为了达到行数指标而继续拆分已经职责清楚的模块。

如果同一批改动同时包含功能变化和结构重构,先在记录中明确“允许变化”和“不允许变化”。任何未列入允许变化的差异都按潜在回归处理。

强制原则

  1. 必须使用 Codex 内置多代理协作工具;不得用外部编排器、额外工作树、独立 Codex task 或外部 Agent 代替。项目开发 Runner 使用 aaalice-dev-sessions 管理的独立命令行窗口,不属于子代理替代方案。
  2. 审查代理和确认代理必须是不同的新代理;任何代理不得确认自己的结论。
  3. 审查默认只读;没有经过反证确认的问题不得进入修复。
  4. 不把文件行数、个人风格、抽象偏好或“可能以后出错”单独当成回归。
  5. 修复代理只能按互不重叠的问题簇并行。涉及同一文件、同一状态机或同一调用链时必须顺序执行。
  6. 主 Agent 必须交叉核对所有子代理输出,不按多数票机械裁决,也不把交付责任转移给子代理。
  7. 每次修复后必须开始新一轮完整审查;修复发生过的轮次不能直接判定通过。
  8. 不设置为了快速结束而存在的轮次上限。遇到外部阻塞、证据不足或连续两轮没有实质进展时停止并如实报告,不伪造成功。

进度记录

为每个任务创建:

text
tool/.tmp/refactor-parity/<task-slug>/progress.md
tool/.tmp/refactor-parity/<task-slug>/workers/

tool/.tmp/ 已被 Git 忽略。首次创建时复制模板,再补全内容:

powershell
$ErrorActionPreference = 'Stop'
$root = 'tool/.tmp/refactor-parity/<task-slug>'
New-Item -ItemType Directory -Force "$root/workers" | Out-Null
Copy-Item '.agents/skills/aaalice-refactor-parity-loop/references/progress-template.md' "$root/progress.md"

记录是审计轨迹,不是事实来源。后续代理必须重新检查代码、diff、测试和运行证据,不能因为 Markdown 中写了“已确认”就接受结论。

修复代理完成后,必须在以下唯一文件写自己的报告,避免并发修改主记录:

text
tool/.tmp/refactor-parity/<task-slug>/workers/round-<N>-<worker-name>.md

报告至少包含:负责的问题 ID、修改文件、保持的契约、验证命令与结果、剩余疑点。主 Agent 将其摘要和实际 diff 核验结果写回 progress.md。

阶段 0:锁定基线与等价性契约

修改前先完成:

  1. 记录仓库路径、当前分支、HEAD、比较基线 SHA 和工作区现有改动。
  2. 基线优先使用用户指定提交;否则使用重构开始前的精确提交。不得默认用 origin/main 覆盖当前功能分支语义。
  3. 如果重构已经发生但尚未提交,使用 git diff 与 git show HEAD:<path> 重建拆分前内容;明确哪些基线证据已经无法获取。
  4. 列出目标文件、调用方、公共类型/方法/回调、Provider、路由、平台桥接、资源和测试。
  5. 建立等价性矩阵,至少包含:
    • 输入、输出、异常和副作用;
    • 状态初始化、更新、销毁、取消和异步竞态;
    • 路由、返回、焦点、键盘、鼠标、触控和手势;
    • 空态、加载、错误、有数据、禁用和并发状态;
    • Windows、Android、桌面窄窗、手机、横屏和平板差异;
    • 文案、本地化键、语义、可访问性和快捷键;
    • 受影响布局的尺寸、顺序、显隐、滚动、Overlay 和 SafeArea。
  6. 运行预计 60 秒内可完成的重构前针对性测试并记录结果。无法运行时记录原因,不能假定基线通过。

如果改动涉及 UI:

  • 优先把关键尺寸和状态固化为确定性 widget/layout test。
  • 只有用户明确要求自动化运行验收时,才加载 aaalice-runtime-verify 获取重构前后的真实截图、交互和日志证据。
  • 没有运行态授权且没有直接覆盖受影响状态的确定性布局测试时,不得声称已经证明像素级或真实运行布局完全一致。

阶段 1:执行行为保持型拆分

  1. 按职责拆分,不顺手改变交互、样式、文案、业务规则或平台能力。
  2. 保持公共入口、调用顺序、Provider 生命周期、Widget key、语义和回调契约,除非等价性矩阵明确允许变化。
  3. 每完成一个可编译边界就运行最小检查;不要积累大量未验证移动后再一次性修错。
  4. 工作区已有改动默认属于用户,不回滚、不覆盖、不格式化无关文件。
  5. 拆分完成后更新拆分清单:旧成员到新文件/新类型的映射,以及有意删除的死代码证据。

阶段 2:多代理独立审查

在当前可用并发槽内优先并行派发 3 个只读子代理;无法同时覆盖的维度在子代理完成后继续派发,范围尽量不重叠:

  1. 功能与 API 契约:调用链、输入输出、回调、副作用、错误和兼容性。
  2. 状态与生命周期:Provider/Controller/State、异步取消、初始化、销毁、导航和资源释放。
  3. UI 与布局:Widget 树、constraints、响应式、Overlay、滚动、焦点、手势、语义和平台差异。
  4. 接线完整性:import/export、生成代码、路由、依赖注入、本地化、资源、测试入口和死代码。
  5. 平台专项(按需):Windows、Android 或原生桥接的等价性。

所有审查代理必须独立比较基线与当前实现,不读取其他代理结论。每条发现必须包含:稳定 ID、基线行为、当前行为、准确路径/行号、可触发条件、影响、置信度和最小验证方式。

以下内容不得单独登记为问题:

  • “文件仍然很长”;
  • “我更喜欢另一种抽象”;
  • 没有调用链或行为证据的假设性风险;
  • 与本次 diff 无关的既有问题;
  • 仅缺少额外测试但没有发现行为差异。

主 Agent 去重后把候选发现写入当前轮次,但此时状态只能是 待确认。

阶段 3:多代理对抗性确认

派发至少 2 个新的只读子代理。将候选发现按调用链或文件簇分配,重要问题至少由两个确认代理独立检查。

确认代理必须先尝试反证:

  • 查找 guard、fallback、最近作用域处理、共享命令入口、测试和框架真实机制;
  • 判断差异是否属于等价性矩阵允许变化;
  • 构造最小反例或针对性测试思路;
  • 区分“当前真实回归”“维护风险”“纯风格偏好”和“误报”。

分类规则:

  • 已确认:存在直接代码、测试、日志或运行证据,且能说明触发路径和行为差异。
  • 已反证:框架机制、现有 guard、测试或调用链证明原结论不成立。
  • 未决:证据不足或结论依赖未验证环境。

未决 不能进入修复。主 Agent必须追加只读检查或最小测试来解决;无法解决时本轮不能通过。

如果独立审查没有发现问题,确认代理仍要执行“清洁轮挑战”:尝试寻找审查遗漏、验证关键等价性矩阵和测试覆盖。只有挑战也没有产生 已确认/未决 问题,才算干净轮候选。

如果审查发现的问题全部被反证,且确认代理没有发现新问题,可以进入完成门禁,不需要为了“做点修改”而制造修复。

阶段 4:分组修复已确认问题

  1. 只修复 已确认 且由本次重构引入的问题。
  2. 按互不重叠的文件和调用链建立修复簇,派发多个修复子代理;强耦合问题由一个代理完整处理。
  3. 每个修复代理获得:问题 ID、基线契约、允许修改文件、禁止范围、预期验证和独立报告路径。
  4. 修复代理不得扩大功能、修改 Changelog、隐藏异常、降低测试或自行宣布整轮通过。
  5. 主 Agent 逐个检查 worker 报告、实际 diff、调用方和测试,解决冲突并执行必要整合。
  6. 修复完成后更新主记录,然后回到阶段 2,启动全新的完整审查轮。

阶段 5:完成门禁

只有同时满足以下条件才能把状态写为 PASS:

  1. 最近一轮在所有相关审查维度上完成了独立审查。
  2. 新确认代理完成了对抗性确认或清洁轮挑战。
  3. 最近一轮没有 已确认 或 未决 问题;候选问题可以为空或全部被可靠反证。
  4. 最近一轮之后没有再修改生产代码;如果修改过,必须重新开轮。
  5. 等价性矩阵全部有代码、测试、日志或运行证据;不适用项有理由。
  6. 针对性测试、相关集成测试和 scoped analyze 通过;既有或环境失败已区分并记录。
  7. 最终 diff 只有任务范围内的结构变化和必要回归修复,没有第二事实来源、静默降级、宽泛吞错、死代码或无关格式化。
  8. UI 重构具备直接布局证据。若既没有对应 widget/layout test,也没有经用户授权的运行态对比,状态必须保持 BLOCKED_LAYOUT_EVIDENCE,不能声称完整布局对齐。

完成时在 progress.md 写明:最终状态、干净轮编号、基线与最终 SHA/diff、验证命令、运行态是否执行、所有已反证问题及依据、剩余非阻塞事项。

最终回复

简洁报告:

  • 拆分范围和新模块边界;
  • 审查轮数、确认轮数和修复问题数量;
  • 最终干净轮及进度记录路径;
  • 实际执行的测试、分析和运行态验证;
  • PASS、BLOCKED_LAYOUT_EVIDENCE 或其他阻塞状态。

没有满足完成门禁时不得使用“完全对齐”“已经成功”或“无回归”。

© Aaalice233, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in .agents/skills/aaalice-refactor-parity-loop of Aaalice233/Aaalice_NAI_Launcher.

  • SKILL.md
  • references/agent-prompts.md
  • references/progress-template.md

Open the folder on GitHubat commit fe0f12b

Compare with similar skills

Aaalice Refactor Parity Loop next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Aaalice Refactor Parity Loop compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Aaalice Refactor Parity Loop this skillAaalice233/Aaalice_NAI_Launcher188—~1.3kAutomated safety check: PassMIT
Codexskills-directory/skill-codex1.5k3 repos~1.8kAutomated safety check: PassMIT
Nexus MapperHaaaiawd/Nexus-skills166—~2.5kAutomated safety check: NotesNone
Refactoring Auditepam/ai-dial-chat504—~4.3kAutomated safety check: PassApache-2.0
Nexus QueryHaaaiawd/Nexus-skills166—~1.1kAutomated safety check: PassNone
Refactortermide/termide172—~2.8kAutomated safety check: PassMIT

Similar skills

  • Codex

    skills-directory/skill-codex

    A skill your agent uses when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automated editing

    1.5k GitHub starsUsed in 3 repos~1.8k tokens
    DevelopmentAuto-check passed
  • Nexus Mapper

    Haaaiawd/Nexus-skills

    Generate a persistent .nexus-map/ knowledge base that lets any AI session instantly understand a codebase's architecture, systems, dependencies, and change hotspots.

    166 GitHub stars~2.5k tokensUpdated 6 mo ago
    DevelopmentAuto-check: notes
  • Refactoring Audit

    epam/ai-dial-chat

    Deep codebase refactoring audit for AI DIAL Chat. An agent skill from epam/ai-dial-chat.

    504 GitHub stars~4.3k tokensUpdated today
    DevelopmentAuto-check passed
  • Nexus Query

    Haaaiawd/Nexus-skills

    Precise, instant code structure queries for active development — answer 'who depends on this interface before I refactor it', 'how many modules break if I change this', 'what is the real impact…

    166 GitHub stars~1.1k tokensUpdated 6 mo ago
    DevelopmentAuto-check passed
  • Refactor

    termide/termide

    Full-workspace code quality analysis and refactoring with validation

    172 GitHub stars~2.8k tokensUpdated today
    DevelopmentAuto-check passed
  • Merge Master

    keybase/client

    A skill your agent uses when periodically merging upstream master commits into a long-running refactor branch where code has moved, been renamed, or restructured — guiding careful per-commit…

    9.3k GitHub stars~1.3k tokensUpdated today
    DevelopmentAuto-check passed

More from Aaalice233/Aaalice_NAI_Launcher

  • Aaalice Dev Sessions

    Aaalice233/Aaalice_NAI_Launcher

    管理 Aaalice NAI Launcher 项目专用的 Windows 与 Android Flutter 开发会话。用户提到 PC热重载、安卓热重载、启动/关闭/重建独立命令行窗口、模拟器启动异常或检查双端会话时使用。

    188 GitHub stars~902 tokensUpdated 3 days ago
    Auto-check passed
  • Aaalice Hot Reload

    Aaalice233/Aaalice_NAI_Launcher

    为 Aaalice NAI Launcher 修改代码后选择并执行 Windows/Android 的 r 热重载、R 热重启或完整重建,并验证结果。已有开发会话运行时,代码修改完成后自动使用;用户要求热重载、热重启、修改后刷新或查看 Flutter 日志时也使用。

    188 GitHub stars~531 tokensUpdated 3 days ago
    Auto-check passed
  • Aaalice Runtime Verify

    Aaalice233/Aaalice_NAI_Launcher

    用户要求自动化运行验收、界面操作、截图检查或双端 UI 测试时,自动启动或复用 Aaalice NAI Launcher 热重载会话,通过项目级 Dart/Flutter MCP 验证应用;Android 系统界面按需使用 ADB。

    188 GitHub stars~963 tokensUpdated 3 days ago
    Auto-check passed
  • Aaalice Launcher Release

    Aaalice233/Aaalice_NAI_Launcher

    为 Aaalice NAI Launcher 准备并发布新版本,包括同步 main、更新版本号、生成并逐项审查 Changelog 材料、撰写用户更新日志、执行发布前验证、提交、创建并推送 v tag。用户提到发布 Launcher 版本、准备 Release、写 CHANGELOG、更新日志或 release notes 时使用。

    188 GitHub stars~1.2k tokensUpdated 3 days ago
    Auto-check passed

Works with

Categories

Questions about Aaalice Refactor Parity Loop

What does Aaalice Refactor Parity Loop do?

对 Aaalice NAI Launcher 的大型文件拆分型重构执行多子代理等价性闭环。当任务目标是把大型业务文件按职责拆分、同时保持功能、交互与布局不变时自动使用;不用于普通重构、功能开发、单点 Bug 修复或常规代码审查。. Aaalice Refactor Parity Loop is an agent skill from Aaalice233/Aaalice_NAI_Launcher.

When should I use Aaalice Refactor Parity Loop?

Aaalice Refactor Parity Loop fits situations like: tasks that involve Refactoring.

How do I install Aaalice Refactor Parity Loop in Claude Code?

Run `npx skills add Aaalice233/Aaalice_NAI_Launcher --skill aaalice-refactor-parity-loop -a claude-code`. Or copy the skill folder (.agents/skills/aaalice-refactor-parity-loop in Aaalice233/Aaalice_NAI_Launcher) into .claude/skills/aaalice-refactor-parity-loop in your project. Claude Code loads it when a task matches its description.

How do I install Aaalice Refactor Parity Loop in Codex?

Run `npx skills add Aaalice233/Aaalice_NAI_Launcher --skill aaalice-refactor-parity-loop -a codex`. Or copy the skill folder (.agents/skills/aaalice-refactor-parity-loop in Aaalice233/Aaalice_NAI_Launcher) into .agents/skills/aaalice-refactor-parity-loop in your project. Codex loads it when a task matches its description.

Can I use Aaalice Refactor Parity Loop in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Aaalice233/Aaalice_NAI_Launcher --skill aaalice-refactor-parity-loop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/aaalice-refactor-parity-loop, .gemini/skills/aaalice-refactor-parity-loop, .github/skills/aaalice-refactor-parity-loop and .opencode/skills/aaalice-refactor-parity-loop in your project.

What does Aaalice Refactor Parity Loop need to run?

Going by SKILL.md and its folder, Aaalice Refactor Parity Loop needs the command-line tools its instructions call (git).

Does Aaalice Refactor Parity Loop access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Aaalice Refactor Parity Loop safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Aaalice Refactor Parity Loop use?

Aaalice Refactor Parity Loop is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Aaalice Refactor Parity Loop use?

About 1.3k tokens (SKILL.md is roughly 5.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.6k tokens, read only when the agent opens those files.

What are the alternatives to Aaalice Refactor Parity Loop?

Skills that share tags, products or a category with Aaalice Refactor Parity Loop: Codex (skills-directory/skill-codex, 1.5k stars), Nexus Mapper (Haaaiawd/Nexus-skills, 166 stars), Refactoring Audit (epam/ai-dial-chat, 504 stars) and Nexus Query (Haaaiawd/Nexus-skills, 166 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Aaalice Refactor Parity Loop?

Aaalice233 (a GitHub user) maintains it in Aaalice233/Aaalice_NAI_Launcher, which has 188 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 5, 2026.

Source: Aaalice233/Aaalice_NAI_Launcher on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.