# SDLC AI Workflow: Claude Code Plugin Enables Full-Cycle Software Development Automation > Explore the saitarrun/sdlc-ai-workflow project, a plugin designed for Claude Code that automates the complete software development lifecycle from requirement planning to operation and maintenance monitoring through 26 professional AI agents, 43 knowledge skills, and 14 commands. - 板块: [Openclaw Llm](https://www.zingnex.cn/en/forum/board/openclaw-llm) - 发布时间: 2026-06-17T00:16:09.000Z - 最近活动: 2026-06-17T00:20:19.507Z - 热度: 163.9 - 关键词: SDLC, AI代理, Claude Code, 软件开发自动化, DevOps, 测试自动化, CI/CD, 软件架构, 敏捷开发, AI辅助开发 - 页面链接: https://www.zingnex.cn/en/forum/thread/sdlc-ai-workflow-claude-code - Canonical: https://www.zingnex.cn/forum/thread/sdlc-ai-workflow-claude-code - Markdown 来源: floors_fallback --- ## Introduction / Main Floor: SDLC AI Workflow: Claude Code Plugin Enables Full-Cycle Software Development Automation Explore the saitarrun/sdlc-ai-workflow project, a plugin designed for Claude Code that automates the complete software development lifecycle from requirement planning to operation and maintenance monitoring through 26 professional AI agents, 43 knowledge skills, and 14 commands. ## Original Author and Source - **Original Author/Maintainer**: saitarrun - **Source Platform**: GitHub - **Original Title**: sdlc-ai-workflow - **Original Link**: https://github.com/saitarrun/sdlc-ai-workflow - **Publication Date**: June 17, 2026 --- ## Project Overview: AI-Driven End-to-End Development Orchestration In modern software development, teams often face a core pain point: **context loss, delays, and errors caused by manual handoffs**. From product managers to architects, then to engineers, testers, and operations personnel, each role transition can lead to information decay. The SDLC AI Workflow project was created to address this issue—it is a plugin designed specifically for Claude Code that automates the complete Software Development Lifecycle (SDLC) from planning to operations through the collaborative work of 26 professional AI agents. The core philosophy of this project is to encode industry best practices into the behavior of AI agents, allowing the entire development process to run smoothly like an assembly line while maintaining high-quality standards. --- ## Six Development Phases and 26 Professional Agents The project divides software development into six distinct phases, each with dedicated AI agents responsible for it: ## Phase 1: Planning and Requirement Analysis This is the starting point of the project, determining the direction of all subsequent work. This phase includes four core agents: - **Product Manager**: Responsible for formulating roadmaps and milestones, using the QUANTS framework to evaluate requirements - **Business Analyst**: Writes user stories and acceptance criteria, applies the INVEST principle to ensure story quality - **Software Architect**: Selects the tech stack, writes Architecture Decision Records (ADR), and draws system blueprints - **Security Architect**: Conducts threat modeling, uses the STRIDE method to identify potential security risks These agents automatically load phase-specific skills; for example, requirement analysis skills inject knowledge of the INVEST standard and QUANTS framework to ensure outputs comply with industry norms. ## Phase 2: Design and Prototyping After requirements are clarified, the design phase transforms concepts into interactive blueprints: - **UX Researcher**: Analyzes user journeys and builds user personas - **UI/UX Designer**: Creates wireframes and design system specifications This phase emphasizes rapid validation of design hypotheses, using disposable prototype code to test key interaction flows and avoid investing too many development resources in the wrong direction. ## Phase 3: Development and Implementation This is the most labor-intensive phase, involving five professional development agents: - **Frontend Engineer**: Responsible for user interfaces and client-side logic - **Backend Engineer**: Builds servers, APIs, and business logic - **Full-Stack Engineer**: End-to-end feature implementation - **Database Engineer**: Designs schemas, migration scripts, and indexing strategies - **Mobile Developer**: iOS/Android native app development Each development agent automatically applies coding standards such as Clean Code, SOLID principles, and DRY (Don't Repeat Yourself) to ensure code quality is guaranteed from the source. ## Phase 4: Testing and Security Audit The quality assurance phase adopts a multi-layered testing strategy: - **QA Manual Tester**: Conducts exploratory testing to find potential issues from the user's perspective - **Automation QA Engineer**: Generates test suites and integrates them into the CI process - **Application Security Engineer**: Performs SAST static analysis, CVE scanning, and OWASP Top 10 checks - **Penetration Tester**: Simulates attack scenarios to test injection vulnerabilities and authentication bypasses The test pyramid principle is automatically applied to ensure a reasonable ratio of unit tests, integration tests, and end-to-end tests, avoiding the accumulation of test debt.