openspec-new-change
A CLI-based tool that uses OpenSpec workflows to scaffold new features or bug fixes with structured documentation and development steps.
Install
mkdir -p .claude/skills/openspec-new-change && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/821" && unzip -o skill.zip -d .claude/skills/openspec-new-change && rm skill.zipInstalls to .claude/skills/openspec-new-change
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
使用实验性的产出物工作流启动一个新的 OpenSpec 变更。当用户想要通过结构化的分步方法创建新功能、修复或修改时使用。Key capabilities
- →Scaffold a new change directory with a specified name
- →Apply a default or user-selected workflow schema
- →Display the status of artifacts within a change
- →Provide instructions for the first ready artifact
- →Infer a kebab-case name from a user description
How it works
The skill prompts for a change description, infers a name, creates a scaffolded change directory using a schema, and then displays the status and instructions for the initial artifact.
Inputs & outputs
When to use openspec-new-change
- →Starting a new feature branch via OpenSpec
- →Generating scaffolding for bug fixes
- →Tracking status of development artifacts
- →Managing specification-driven development changes
About this skill
使用实验性的产出物驱动方法启动新变更。
输入:用户的请求应当包含变更名称(kebab-case)或对想要构建内容的描述。
步骤
-
如果没有提供明确的输入,询问用户想要构建什么
使用 AskUserQuestion Tool(开放式,无预设选项)询问:
"您想要处理什么变更?请描述您想要构建或修复的内容。"
根据他们的描述,推导出一个 kebab-case 名称(例如:"add user authentication" →
add-user-auth)。重要提示:在不了解用户想要构建什么的情况下,请勿继续。
-
确定工作流 Schema
除非用户明确要求不同的工作流,否则使用默认 Schema(省略
--schema)。仅当用户提到以下内容时才使用不同的模式:
- 特定模式名称 → 使用
--schema <name> - "显示工作流" 或 "有哪些工作流" → 运行
openspec-cn schemas --json并让他们选择
否则:省略
--schema以使用默认值。 - 特定模式名称 → 使用
-
创建变更目录
openspec-cn new change "<name>"仅当用户请求特定工作流时才添加
--schema <name>。 这将在openspec/changes/<name>/下使用所选 Schema 创建一个脚手架变更。 -
显示产出物状态
openspec-cn status --change "<name>"这会显示哪些产出物需要创建,以及哪些已就绪(依赖项已满足)。
-
获取第一个产出物的指令 第一个产出物取决于所使用的 schema(例如:spec-driven 通常先生成
proposal)。 检查 status 输出,找到第一个状态为 "ready" 的产出物。openspec-cn instructions <first-artifact-id> --change "<name>"这会输出创建第一个产出物所需的模板和上下文。
-
停止并等待用户指示
输出
完成上述步骤后,进行总结:
- 变更名称和位置
- 正在使用的 Schema/工作流及其产出物顺序
- 当前状态(0/N 个产出物已完成)
- 第一个产出物的模板
- 提示:"准备好创建第一个产出物了吗?请描述此变更的内容,我将为您起草,或者要求我继续。"
护栏
- 不要立即创建任何产出物 —— 仅显示指令
- 不要跳过显示第一个产出物模板的步骤
- 如果名称无效(非 kebab-case),请求有效的名称
- 如果同名变更已存在,建议继续处理该变更
- 如果使用非默认工作流,请传递 --schema
When not to use it
- →When the user wants to immediately create artifacts
- →When the user does not want to use a structured, step-by-step method
Prerequisites
Limitations
- →It does not immediately create any artifacts
- →It requires a valid kebab-case name for the change
- →It will not proceed without knowing what the user wants to build
How it compares
This skill automates the initial setup and guidance for OpenSpec changes, unlike manually creating directories and looking up schema requirements.
Compared to similar skills
openspec-new-change side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| openspec-new-change (this skill) | 4 | 6mo | Review | Beginner |
| openspec-onboard | 10 | 6mo | Review | Beginner |
| workflow-orchestration-patterns | 10 | 2mo | No flags | Advanced |
| meta-automation-architect | 7 | 8mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by studyzy
View all by studyzy →You might also like
openspec-onboard
studyzy
Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.
workflow-orchestration-patterns
wshobson
Design durable workflows with Temporal for distributed systems. Covers workflow vs activity separation, saga patterns, state management, and determinism constraints. Use when building long-running processes, distributed transactions, or microservice orchestration.
meta-automation-architect
comzine
Use when user wants to set up comprehensive automation for their project. Generates custom subagents, skills, commands, and hooks tailored to project needs. Creates a multi-agent system with robust communication protocol.
hive-mind-advanced
ruvnet
Advanced Hive Mind collective intelligence system for queen-led multi-agent coordination with consensus mechanisms and persistent memory
router-first-architecture
parcadei
Router-First Architecture
mcp-scripts
parcadei
MCP Script Rules