Skills Plugins MCP Prompt Model 博客 我的中心

spec-driven

Spec-driven feature work with traceable requirements, design, tasks, audit, and validation. Use when planning, designing, implementing, or validating a feature. Not for unknown-bug diagnosis, standalone product or technical documents, PR or commit mechanics, or backlog tracking.

DeepseekModel 官方收录技能 质量 良好 · 64 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=adeonir-agent-skills-skills-engineering-spec-driven-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name spec-driven description Spec-driven feature work with traceable requirements, design, tasks, audit, and validation. Use when planning, designing, implementing, or validating a feature. Not for unknown-bug diagnosis, standalone product or technical documents, PR or commit mechanics, or backlog tracking. argument-hint [T-N] | [T-N..T-M] | [S-N] | [S-N..S-M] | [W-N] | [W-N..W-M] allowed-tools Bash(git:*) Bash(python3:*) Read Write Edit Grep Glob Task Spec-Driven Development Feature development in phases. Light by default; weight only where the change pays for it. Triggers Specify ("plan feature", "spec this", "from PRD", "modify feature", "discuss how to build") → specify.md Design ("design this feature", "technical design", "plan the build") → design.md Tasks ("create tasks", "break into tasks", "task breakdown") → tasks.md Implement ("implement task T-1", "implement T-1 to T-4", "implement slice S-1", "implement wave W-1", "execute tasks", "implement everything") → implement.md Audit ("audit feature", "validate goals", "verify before PR") → audit.md Validate / UAT ("run UAT", "manual testing", "validate flows") → validate.md Archive ("archive feature", "archive this spec") → archive.md Workflow specify → design → tasks → implement → [validate] → [audit] → [archive] └────────┴────────┴──────────┴──────────┴ a mechanical change skips all of this: one-liner → branch → implement inline Specify's triage decides the path: a mechanical change with zero load-bearing decisions becomes a one-liner straight to inline implement on its own branch, and a prompt carrying outcomes that ship separately becomes one feature per outcome. Everything else produces the artifacts and runs the phases in turn. Verify is mental, per task, inside implement — never a user phase. Validate and audit are optional. Archive is manual housekeeping for a feature in any state, never automatic or suggested.
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。