Token导航 LogoToken导航TokenDH.com
研究检索external-servicegithub未标认证来源可访问许可证需确认审计提醒

create-output-format创建输出格式

Agent Skill

create-output-format 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

808

周安装

34

GitHub Stars

15

下载量

283
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:create-output-format(创建输出格式)
来源仓库:https://github.com/leeovery/claude-technical-workflows
仓库路径:skills/create-output-format
安装命令:
npx skills add https://github.com/leeovery/claude-technical-workflows --skill create-output-format
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/leeovery/claude-technical-workflows --skill create-output-format

简介

为工作流规划流程创建新的输出格式适配器,每个格式对应五个关注点文件。

  • 需调研目标工具的存储机制和交互方式,建立任务持久化和结果返回规范。
  • 支持多种输出格式扩展,包括 JSON、YAML 和自定义 schema 验证结构。
  • 使用前应确认目标系统是否支持所选格式,避免因兼容性问题导致数据丢失。
  • create-output-format 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Create Output Format

Scaffold a new output format adapter for the workflow-planning-process workflow. Each format adapter is a directory of 5 files, one per concern.

Step 1: Gather Information

Before writing anything, understand the tool. Ask the user for:

  1. What is the tool/system called?
  2. Documentation links — official docs, API references, MCP server docs, anything relevant

If the user provided these upfront, skip straight to research.

Research

Fetch and read all provided documentation using WebFetch. From the docs, establish:

  • How tasks are stored (API, database, files, etc.)
  • How to interact with it (MCP server, REST API, CLI tool, filesystem)
  • How to create, read, update, and query tasks
  • What task properties are supported — status values, priority levels, labels/tags, estimation
  • Whether it supports blocking/dependency relationships (within a project and across projects)
  • How concepts map — what represents a project, a phase, a task, a dependency?
  • What setup or configuration is required
  • Benefits and trade-offs vs simpler approaches
  • Any constraints or limitations

Clarify Gaps

Present what you've learned as a summary and ask the user to confirm or correct. Use AskUserQuestion to clarify anything the documentation didn't cover or left ambiguous — motivation for choosing this format, preferred interface if multiple exist, setup specifics, etc.

Suggest a kebab-case format key based on the tool name and confirm with the user.

Do not proceed until the user confirms your understanding.

Step 2: Understand the Contract

Read references/contract.md — this defines the 5-file interface every format must implement.

Step 3: Create the Format Directory

Create the directory at:

skills/workflow-planning-process/references/output-formats/{format-key}/

Step 4: Write the Files

Using the information gathered in Step 1, write each of the 5 required files. Use the scaffolding templates from references/scaffolding/ as structural guides:

TemplateCreates
about.md{format}/about.md
authoring.md{format}/authoring.md
reading.md{format}/reading.md
updating.md{format}/updating.md
graph.md{format}/graph.md

For each file:

  1. Start from the scaffolding template structure
  2. Replace all {placeholder} tokens with format-specific content from your gathered information
  3. Remove template guidance comments (lines starting with <!-- -->)
  4. Include concrete commands, API calls, or MCP operations — not vague descriptions

Step 5: Register the Format

Add an entry to skills/workflow-planning-process/references/output-formats.md following the existing pattern:

### {Format Name}
format: `{format-key}`

adapter: [{format-key}/](output-formats/{format-key}/)

{One-line description of the format.}

- **Pros**: ...
- **Cons**: ...
- **Best for**: ...

Step 6: Validate

Verify:

  • Directory contains exactly 5 files: about.md, authoring.md, reading.md, updating.md, graph.md
  • All {placeholder} tokens have been replaced
  • About.md includes a structure mapping table
  • Authoring.md documents task properties: status, phase grouping, labels (NOT priority or dependencies)
  • Authoring.md includes a complete task creation example
  • Reading.md explains next-task ordering using status, priority, dependencies, and phase
  • Updating.md covers all status transitions and how to modify task properties
  • Graph.md covers priority levels and adding/removing dependencies
  • Format is registered in output-formats.md

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

34.6%
按下载量换算98

Claude

31.33%
按下载量换算89

Cursor

20.13%
按下载量换算57

Gemini CLI

9.93%
按下载量换算28

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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