Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计提醒

codingcoding 搜索

Agent Skill

coding 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

1,397

周安装

60

GitHub Stars

35

下载量

490
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:coding(coding 搜索)
来源仓库:https://github.com/bout3fiddy/agents
仓库路径:skills/coding
安装命令:
npx skills add https://github.com/bout3fiddy/agents --skill coding
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/bout3fiddy/agents --skill coding

简介

coding 用于代码实现、审查和仓库代码变更技术指导,支持多种编程语言和技术栈。

  • 适用场景包括代码编辑、质量评审、PR 反馈循环和基础设施实施,确保代码质量和一致性。
  • 核心能力包括技术指导和架构建议,避免将规划或 AGENTS 架构任务路由到其他技能。
  • 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。
  • coding 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Coding (Implementation + Review)

Use this skill when the primary intent is implementation, bug fixing, refactoring, code-quality review, or technical guidance for repository code changes.

Scope and routing

  • Use this skill for code edits, code-quality reviews, PR feedback loops, and infra/platform implementation tasks.
  • If the request is planning/spec/lifecycle management, hand off to planning.
  • If the request is AGENTS architecture housekeeping, hand off to housekeeping.
  • If the request is under skills/ or edits SKILL.md, hand off to skill-creator.
  • For frontend framework guidance, use SolidJS references only (skills/coding/references/solidjs/...).

Mandatory smell baseline (always for code changes)

For implementation/bugfix/refactor/review operations, always open:

  • skills/coding/references/code-smells/index.md
  • skills/coding/references/code-smells/detection-signals.md

Then open specific smell files for detailed refactor guidance when detection signals match. When writing new code (not just editing), err on the side of reading more smell refs — it is cheaper to check and dismiss than to miss a smell.

Core workflow

  1. Confirm scope, constraints, and acceptance criteria.
  2. Load mandatory smell baseline refs.
  3. Load domain refs relevant to the task.
  4. Read only necessary code paths.
  5. Implement minimal focused changes.
  6. Self-review: check your own output against detection signals. Fix violations before presenting — especially Duplicate Code, Speculative Generality, and Long Method, which are easy to introduce organically.
  7. Validate with targeted checks/tests.
  8. Summarize changes, validations, and remaining risks.

Domain reference triggers (open when clearly relevant)

  • Code smell / maintainability / quality diagnostics:

- skills/coding/references/code-smells/index.md - skills/coding/references/code-smells/smells/index.md

  • Work-package refactoring execution:

- skills/coding/references/refactoring/index.md

  • SolidJS implementation/performance:

- skills/coding/references/solidjs/index.md - skills/coding/references/solidjs/rules/index.md

  • Infra / deploy / platform ops:

- skills/coding/references/platform-engineering/index.md

  • Auth / credentials / secret handling:

- skills/coding/references/secrets-and-auth-guardrails.md

  • PR review bot loop / CI failure remediation:

- skills/coding/references/gh-pr-review-fix.md

  • JS/TS runtime and toolchain:

- skills/coding/references/bun.md

Quality rules

  • Prefer hard cutovers over fallback-first compatibility branches.
  • Avoid over-defensive code that obscures normal control flow.
  • Preserve existing architecture unless the task explicitly asks for structural change.
  • Add/update tests when behavior changes.
  • Keep edits small, cohesive, and traceable to the request.

Output expectations

  • Explain decisions and key trade-offs.
  • List changed files.
  • Report validations run (or why skipped).
  • Call out unresolved risks and follow-up actions.

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

35.64%
按下载量换算175

Claude

30.96%
按下载量换算152

Cursor

19.67%
按下载量换算96

Gemini CLI

8.86%
按下载量换算43

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。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