Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问clear审计未展示

writing-commands写命令

Agent Skill

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

总安装

247

周安装

10

GitHub Stars

公开资料未说明

下载量

78
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

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:writing-commands(写命令)
来源仓库:https://github.com/ian-pascoe/dotfiles
仓库路径:skills/writing-commands
安装命令:
npx skills add ian-pascoe/dotfiles --skill "writing-commands"
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。

AgentSkills.tonpx skills
npx skills add ian-pascoe/dotfiles --skill "writing-commands"

简介

辅助文档、README 和内容稿件的整理与改写,提升可读性。

  • 支持结构梳理、术语统一和链接检查等文本优化任务。
  • 通过 github 安装,结合实际项目路径调整使用方式。
  • 对外文案需控制语气,避免过度承诺或夸大功能。
  • writing-commands 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

name
writing-commands
description
Use when creating, editing, or reviewing custom OpenCode commands. Covers frontmatter options, prompt templating with arguments and shell output, file references, and agent/model configuration.

Writing Commands

Overview

Commands are custom prompts triggered via /command-name in the TUI. They automate repetitive tasks with configurable agents, models, and dynamic prompt templating.

When to Use

  • Creating a new custom command for repetitive workflows
  • Adding argument handling or shell output injection to commands
  • Configuring which agent or model handles a command
  • Converting inline prompts to reusable commands

Don't use for: one-off prompts, complex multi-step workflows (use agents), or logic that needs code execution.

File Locations

.opencode/command/<name>.md         # Project-local
~/.config/opencode/command/<name>.md  # Global/personal

File name becomes the command name (e.g., test.md -> /test).

Frontmatter Options

---
description: Brief description shown in TUI
agent: build # Optional: agent to execute (default: current)
model: anthropic/... # Optional: override default model
subtask: true # Optional: force subagent invocation
---
OptionRequiredDescription
descriptionNoShown in TUI command list
agentNoAgent to execute (defaults to current)
modelNoOverride model for this command
subtaskNoForce subagent invocation (isolates context)
templateYes\*Prompt text (JSON config only, body in markdown)

\*In markdown files, the body after frontmatter IS the template.

Prompt Templating

Arguments

PlaceholderDescription
$ARGUMENTSAll arguments as single string
$1, $2...Positional arguments
Create a React component named $1 in $2 directory.

Usage: /component Button src/components

Shell Output

Inject command output with backtick syntax:

Here are the current test results:
!`npm test`

Based on these results, suggest improvements.

Commands run from project root.

File References

Include file contents with @path:

Review the component in @src/components/Button.tsx.
Check for performance issues.

Quick Reference

PatternExample
Simple command/test -> runs test suite
With args/component Button -> $1 = Button
Shell injection!git log -5`` -> injects git output
File inclusion@src/index.ts -> includes file content
Subagent isolatedsubtask: true -> runs in isolated context

JSON Config Alternative

Commands can also be defined in opencode.jsonc:

{
  "command": {
    "test": {
      "template": "Run tests with coverage...",
      "description": "Run tests with coverage",
      "agent": "build",
      "model": "anthropic/claude-3-5-sonnet-20241022",
    },
  },
}

Common Mistakes

MistakeFix
Missing descriptionAlways add for TUI discoverability
Using $ARGUMENTS with positionalPick one style: $ARGUMENTS OR $1, $2
Shell commands with side effectsOnly use read-only commands in prompts
Overriding built-in without intentBuilt-ins: /init, /undo, /redo, /share, /help
Forgetting quotes for JSON args/cmd "{ \"key\": \"value\" }"

Validation Checklist

  • [ ] File name is lowercase, hyphen-separated
  • [ ] Description is concise and descriptive
  • [ ] Template uses correct placeholder syntax
  • [ ] Shell commands are read-only (no mutations)
  • [ ] File references use correct @path syntax
  • [ ] Agent exists if specified

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

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

OpenCode

25.46%
按下载量换算20

Claude Code

24.99%
按下载量换算19

windsurf

15.88%
按下载量换算12

Codex

13.23%
按下载量换算10

Antigravity

6.83%
按下载量换算5

Gemini CLI

3.65%
按下载量换算3

安全审计

暂无安全审计结果可展示。

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add ian-pascoe/dotfiles --skill "writing-commands" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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