by TsCarpe
AI-native SDLC skills for Claude Code: requirement intake & audit, adversarial review gate, AI test orchestration, decision re-check, release config audit
# Add to your Claude Code skills
git clone https://github.com/TsCarpe/claude-sdlc-skillsGuides for using ai agents skills like claude-sdlc-skills.
Last scanned: 9/22/2026
{
"issues": [
{
"file": "skills/sdlc-gate/SKILL.md",
"line": 11,
"type": "dangerous-command",
"message": "Dangerous command (writes to Claude config): \">/.claude/skills/sdlc-gate`,全局安装为 `~/.claude/\"",
"severity": "medium"
},
{
"file": "skills/sdlc-test/SKILL.md",
"line": 58,
"type": "dangerous-command",
"message": "Dangerous command (writes to Claude config): \">/.claude/skills/sdlc-test`,全局安装为 `~/.claude/\"",
"severity": "medium"
}
],
"status": "PASSED",
"scannedAt": "2026-09-22T09:04:01.558Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how claude-sdlc-skills compares with popular alternatives.
claude-sdlc-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by TsCarpe. AI-native SDLC skills for Claude Code: requirement intake & audit, adversarial review gate, AI test orchestration, decision re-check, release config audit. It has 63 GitHub stars.
Yes. claude-sdlc-skills passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/TsCarpe/claude-sdlc-skills" and add it to your Claude Code skills directory (see the Installation section above).
claude-sdlc-skills is primarily written in Python. It is open-source under TsCarpe on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh claude-sdlc-skills against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
AI-native SDLC skills for Claude Code(其他兼容 agent 经 skills.sh 亦可安装)——把 SDLC 里最耗人的重复劳动交给 AI:需求梳理、技术设计、测试用例与浏览器执行、对抗评审、发版审计;红线由脚本机械拦截;人只在两处关口裁决。不是 prompt 合集,是带关卡、留档、降级路径的完整工作流。
文档以中文为主,skill 对中英文触发语均响应;在真实 Java 项目上端到端跑通后开源。
npx skills add TsCarpe/claude-sdlc-skills # 一行安装;sdlc-intent 零依赖,装完 5 分钟可试
AI 写代码已经够快,瓶颈移到了需求理解、测试与评审——这些环节要么仍靠人工硬扛,要么让 AI 自由发挥但不可信。这套 skills 用三条纪律回答:
flowchart LR
A[PRD 输入] --> B[sdlc-intent<br/>梳理 + 体检]
B --> C1[技术设计 sdlc-design]
B -.互不阅读.-> C2[测试用例<br/>sdlc-test cases]
C1 --> D[sdlc-gate<br/>分档对抗评审 A=4/B=2/C=自查]
C2 --> D
D --> E{{人 · 逐条裁决放行}}
E --> F[切片开发<br/>大需求走三段式]
F --> G[sdlc-test<br/>静态比对 → 浏览器执行 → 报告]
G --> H{{人 · 复验缺陷}}
DB[sdlc-doubt<br/>决策对抗复查] -. 任意阶段旁路 .-> F
CR[sdlc-config-review<br/>发版配置扫描] -. 发版前旁路 .-> G
六边形 = 人工关口,只有两处;旁路机制随时可插。带真实产物片段的 6 步走查见 example-walkthrough。
| skill | 阶段 | 干什么 · 触发示例 |
|---|---|---|
| sdlc-intent | 需求进来时 | 梳理 PRD 成结构化 digest(角色/概念/功能地图/流程/状态机),再按七层缺陷体检产出分级 issue 报告 + PM 澄清清单。「帮我梳理这个需求文档,完成后继续体检」 |
| sdlc-design | 设计期 | digest 确认后产 design.md:备选方案 / 接口契约读写分级 / 公共资产归属 / D 表(C 约束·H 假设)/ 实施切分约束。/sdlc-design <需求名> |
| sdlc-test | 用例与测试期 | 用例生成 → 静态代码一致性比对 → test 环境浏览器执行 → 报告;通过用例资产化为 Playwright spec,回归交给 runner。/sdlc-test cases|dev|static|exec|spec|report <需求名> |
| sdlc-gate | 设计+用例定稿后 | 按风险分级分档对抗评审(A=4 / B=2 个全新上下文子代理,C=主会话自查)产出 issue 清单,人工逐条裁决后放行开发与测试执行。/sdlc-gate <需求名> |
| sdlc-doubt | 旁路 · 任意阶段 | 对刚做出的非平凡技术决策发起对抗式独立复查。「这个判断我不放心,帮我质疑一下」 |
| sdlc-config-review | 旁路 · 发版前 | 扫描分支 diff,提取发版前需人工处理的事项:Apollo/Nacos 配置 Key、xxl-job 任务、MQ 订阅、知会项。「梳理上线配置清单」 |
| sdlc-guardrails | 横切 · 写入瞬间 | PostToolUse hook 按项目 guardrails.yaml 机械拦截违规写入,pre-commit 兜底 + 存量基线扫描。「给这个项目接红线拦截」 |
| sdlc-sync | 横切 · 产物出口/入口 | 本地 markdown 产物推上团队平台、拉回 PM 澄清答复(平台本体在独立仓 sdlc-platform)。「把三件套推上平台 / 拉一下 PM 的答复」 |
贯穿全链路的关键机制:
第 1 步 · 安装(二选一):
### 方式一:skills CLI(推荐)
npx skills add TsCarpe/claude-sdlc-skills # 装到当前项目 .claude/skills/
npx skills add TsCarpe/claude-sdlc-skills -g # 装到全局 ~/.claude/skills/
npx skills update <skill名> -g # 更新单个;跨文件迁移的版本可能报 Failed to update,remove 后重新 add 即可(见 faq)
### 方式二:Claude Code 插件市场
/plugin marketplace add TsCarpe/claude-sdlc-skills
/plugin install claude-sdlc-skills@claude-sdlc-skills
第 2 步 · 装完先试这句(零外部依赖)——对一个真实需求文档(飞书链接或本地 markdown)说:
帮我梳理这个需求文档 <链接或路径>,完成后继续体检
第 3 步 · 想吃满全套再配:sdlc-guardrails 红线拦截、回归 runner、测试环境文件都在项目侧配置,见 docs/project-setup.md。注意 sdlc-guardrails 装完不会自动生效——需按三步接入挂 hook 并写规则文件(否则写入拦截为 no-op,违规不拦也不提示)。
按依赖重量逐级上,每一级都独立可用(论证见 workflow.md §7.3):
功能点 ≥5 / 涉及表结构变更 / 跨业务域 / AI 判定为大型且用户确认——命中任一就不整块开发(≈ 需求风险分级 A 级),拆成三层金字塔:顶层 parent 做框架,中层骨架 child 先建公共资产,底层业务 child 逐功能点切片。
用三次小确认替代一次看不完的大确认,每次确认的产物粒度都是人能逐行核对的清单。方法论全文:docs/large-req-playbook.md(框架无关,任意任务框架可复刻)+ 三份 产物模板;分级全貌见 sdlc-workflow-landscape。
全部文档带导读地图与三条阅读路径:
sdlc-platform(Spring Boot + Vue + SQLite),skill 源在本仓 skills/sdlc-sync/站在这些来源的肩膀上:addyosmani/agent-skills(纪律强制)、Claude Agent Skills 官方最佳实践(skill 工程规范)、AI-Native SDLC Playbook(工件链思想)。
MIT · 版本兼容:已在 Claude Code + skills CLI(2026-09 版)验证;skill 遵循 Agent Skills 开放标准,其他兼容 agent(Cursor 等)经 skills.sh 亦可安装。更新日志:CHANGELOG.md(最新 v0.16.0)
感谢 LINUX DO 社区的讨论与反馈:https://linux.do