Token导航 LogoToken导航TokenDH.com
前端设计操作浏览器github未标认证来源可访问许可证需确认审计通过

maintainable-typescriptmaintainable TypeScript 命令行

Agent Skill

maintainable-typescript 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

192

周安装

8

GitHub Stars

7

下载量

64
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

2

许可证

unknown

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:maintainable-typescript(maintainable TypeScript 命令行)
来源仓库:https://github.com/miguelspizza/skills
仓库路径:skills/maintainable-typescript
安装命令:
npx skills add https://github.com/miguelspizza/skills --skill maintainable-typescript
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/miguelspizza/skills --skill maintainable-typescript

简介

maintainable-typescript 用于处理 GitHub 仓库和协作信息。

  • 适合围绕仓库状态、代码变更或协作事项进行整理。
  • 可结合来源仓库和原始 README 核验具体用法。
  • 安装前应确认权限范围、维护状态及是否触发联网或文件读写。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Maintainable TypeScript

Use this skill when the project needs maintainability doctrine, not just local code changes.

Layout

  • doctrine/ contains the portable rules and supporting guidance that should hold across strict TypeScript repos.
  • stack/ contains stack-specific doctrine for the opinionated Vite+ / Drizzle / oRPC / Cloudflare setup.
  • prompts/ contains reusable prompt text for cleanup and review tasks when you need to steer another model away from wrapper-heavy refactors.
  • scripts/ contains runnable TypeScript-repo audit helpers for dead code, duplicate code, and import-boundary problems in the current project.
  • assets/tooling-templates/ contains copyable config templates for target repos.

Reading order

Do not read the whole skill directory by default.

  1. Read this file first.
  2. Decide whether the task needs only portable rules or the full house stack.
  3. Load only the doctrine files relevant to the task.
  4. Treat the rest of the skill as reference material, not required context.

Default portable backbone:

If the repo matches the house stack, read stack/start-here.md before any stack-specific files.

Task Router

Use the smallest relevant set.

Cleanup, deletion, and refactor

Package boundaries and shared runtime code

API, schemas, and OpenAPI

Types, constants, and documentation

Testing and high-risk logic

Frontend and React state

Toolchain, dependencies, and database workflow

Full doctrine review or editing this skill itself

  • read all of doctrine/
  • read all of stack/
  • use the bundled verification scripts before finishing

Audit workflow

When the task is cleanup or review, resolve the skill directory first and then run:

skill_dir="<path-to-this-skill>"
bash "$skill_dir/scripts/audit-typescript-repo.sh" .

Treat audit output as signal, not authority. Check real usage before deleting API surface or collapsing a pattern.

If the target repo is Vite+, use vp for the normal toolchain entrypoint: vp lint, vp test, vp fmt, vp pack, vp add, and vp dlx.

Defaults

  • Code is cheap. Structure is expensive.
  • Preserve contracts, tests, and invariants, not stale decomposition.
  • Prefer deletion over shims.
  • Prefer stable subsystem files over one-helper-per-file trees.
  • Prefer derived types and schemas over handwritten duplicates.
  • Prefer durable tests that attach regressions to real product boundaries.
  • Prefer slice integration tests over internally mocked unit tests.
  • Prefer mocking only external systems you do not control.
  • Prefer contract-gated synthetic fixtures over handwritten unchecked JSON.
  • Prefer assertions on rendered output, HTTP behavior, persisted state, and published artifacts over helper call order.
  • Prefer mature tooling for dead code, duplication, and dependency boundaries over manual inspection.
  • Prefer making the codebase more coherent now over promising to clean it up later.

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

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

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

34.89%
按下载量换算22

Claude

27.14%
按下载量换算17

Cursor

19.61%
按下载量换算13

Gemini CLI

9.76%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills