Sky-Cube/fullflow-dev-agent
Full-chain software development agent system for Claude Code — 6 master agents, 25 sub-agents & 96 atomic skills covering product, dev, QA, testing, ops & security.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit Sky-Cube/fullflow-dev-agent/.claude ./rig-fullflow-dev-agent # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/acceptance-criteria .claude/skills/acceptance-criteria $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/agent-orchestration .claude/skills/agent-orchestration $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/ai-architecture-design .claude/skills/ai-architecture-design $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/ai-code-review .claude/skills/ai-code-review $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/ai-code-security-audit .claude/skills/ai-code-security-audit $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/ai-interactive-components .claude/skills/ai-interactive-components $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/answer-accuracy-eval .claude/skills/answer-accuracy-eval $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/api-automation .claude/skills/api-automation $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/api-gateway-config .claude/skills/api-gateway-config $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/api-protocol-definition .claude/skills/api-protocol-definition $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/architecture-compliance .claude/skills/architecture-compliance $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/auto-scaling-config .claude/skills/auto-scaling-config $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/baseline-security-check .claude/skills/baseline-security-check $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/bottleneck-analysis .claude/skills/bottleneck-analysis $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/business-scenario-testing .claude/skills/business-scenario-testing $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/cache-strategy .claude/skills/cache-strategy $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/code-defect-review .claude/skills/code-defect-review $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/code-generation .claude/skills/code-generation $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/code-standard-check .claude/skills/code-standard-check $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/competitor-analysis .claude/skills/competitor-analysis $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/complexity-metrics .claude/skills/complexity-metrics $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/compliance-gap-assessment .claude/skills/compliance-gap-assessment $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/component-library .claude/skills/component-library $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/container-security-scan .claude/skills/container-security-scan $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/cost-optimization .claude/skills/cost-optimization $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/coverage-metrics .claude/skills/coverage-metrics $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/data-encryption .claude/skills/data-encryption $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/data-leak-risk-assessment .claude/skills/data-leak-risk-assessment $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/data-masking .claude/skills/data-masking $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/database-modeling .claude/skills/database-modeling $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/debt-identification-grading .claude/skills/debt-identification-grading $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/debt-repayment-priority .claude/skills/debt-repayment-priority $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/defect-grading-report .claude/skills/defect-grading-report $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/dependency-analysis .claude/skills/dependency-analysis $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/dependency-scan .claude/skills/dependency-scan $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/design-pattern-evaluation .claude/skills/design-pattern-evaluation $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/design-walkthrough .claude/skills/design-walkthrough $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/duplicate-code-detection .claude/skills/duplicate-code-detection $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/environment-init .claude/skills/environment-init $ npx degit Sky-Cube/fullflow-dev-agent/.claude/skills/figma-design-generation .claude/skills/figma-design-generation
$ curl -fsSL --create-dirs -o .claude/agents/dev-ai-engineering-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-ai-engineering-sub.md $ curl -fsSL --create-dirs -o .claude/agents/dev-architecture-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-architecture-sub.md $ curl -fsSL --create-dirs -o .claude/agents/dev-backend-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-backend-sub.md $ curl -fsSL --create-dirs -o .claude/agents/dev-frontend-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-frontend-sub.md $ curl -fsSL --create-dirs -o .claude/agents/dev-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-master.md $ curl -fsSL --create-dirs -o .claude/agents/dev-middleware-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/dev-middleware-sub.md $ curl -fsSL --create-dirs -o .claude/agents/ops-cicd-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/ops-cicd-sub.md $ curl -fsSL --create-dirs -o .claude/agents/ops-cost-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/ops-cost-sub.md $ curl -fsSL --create-dirs -o .claude/agents/ops-deploy-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/ops-deploy-sub.md $ curl -fsSL --create-dirs -o .claude/agents/ops-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/ops-master.md $ curl -fsSL --create-dirs -o .claude/agents/ops-monitor-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/ops-monitor-sub.md $ curl -fsSL --create-dirs -o .claude/agents/product-backlog-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/product-backlog-sub.md $ curl -fsSL --create-dirs -o .claude/agents/product-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/product-master.md $ curl -fsSL --create-dirs -o .claude/agents/product-prd-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/product-prd-sub.md $ curl -fsSL --create-dirs -o .claude/agents/product-prototype-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/product-prototype-sub.md $ curl -fsSL --create-dirs -o .claude/agents/product-requirement-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/product-requirement-sub.md $ curl -fsSL --create-dirs -o .claude/agents/qa-architecture-review-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/qa-architecture-review-sub.md $ curl -fsSL --create-dirs -o .claude/agents/qa-code-review-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/qa-code-review-sub.md $ curl -fsSL --create-dirs -o .claude/agents/qa-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/qa-master.md $ curl -fsSL --create-dirs -o .claude/agents/qa-metrics-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/qa-metrics-sub.md $ curl -fsSL --create-dirs -o .claude/agents/qa-tech-debt-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/qa-tech-debt-sub.md $ curl -fsSL --create-dirs -o .claude/agents/sec-code-audit-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/sec-code-audit-sub.md $ curl -fsSL --create-dirs -o .claude/agents/sec-compliance-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/sec-compliance-sub.md $ curl -fsSL --create-dirs -o .claude/agents/sec-data-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/sec-data-sub.md $ curl -fsSL --create-dirs -o .claude/agents/sec-infra-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/sec-infra-sub.md $ curl -fsSL --create-dirs -o .claude/agents/sec-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/sec-master.md $ curl -fsSL --create-dirs -o .claude/agents/test-ai-special-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/test-ai-special-sub.md $ curl -fsSL --create-dirs -o .claude/agents/test-automation-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/test-automation-sub.md $ curl -fsSL --create-dirs -o .claude/agents/test-functional-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/test-functional-sub.md $ curl -fsSL --create-dirs -o .claude/agents/test-master.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/test-master.md $ curl -fsSL --create-dirs -o .claude/agents/test-performance-sub.md https://raw.githubusercontent.com/Sky-Cube/fullflow-dev-agent/main/.claude/agents/test-performance-sub.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (96)
验收标准定义Skill智能Agent编排SkillAI应用架构设计SkillAI代码专项审查SkillAI代码安全审计SkillAI交互组件开发Skill回答准确性评测Skill接口自动化脚本开发SkillAPI网关配置Skill接口协议定义Skill架构合规性校验Skill弹性伸缩配置Skill基线安全检测Skill性能瓶颈分析Skill业务场景测试执行Skill缓存策略设计实现Skill代码逻辑缺陷审查Skill多语言代码生成Skill编码规范检查Skill竞品与市场分析Skill代码复杂度计算Skill合规差距评估Skill组件库封装Skill容器安全扫描Skill成本优化方案Skill测试覆盖率统计Skill传输存储加密Skill泄露风险评估Skill数据脱敏方案Skill数据库建模Skill债务识别分级Skill偿还优先级排序Skill缺陷分级报告Skill依赖关系分析Skill组件漏洞扫描Skill设计模式评估Skill设计走查Skill重复代码检测Skill环境初始化SkillFigma设计稿生成Skill前端工程化配置Skill全栈架构选型Skill功能规格说明Skill幻觉检测Skill硬编码信息检测Skill镜像构建打包Skill信息架构设计Skill交互流程设计Skill容器编排部署Skill权限最小化校验Skill大模型服务封装Skill压测场景设计Skill压测脚本编写执行Skill日志分析定位Skill监控指标配置Skill消息队列规划开发Skill多环境配置管理Skill多端页面生成Skill网络策略配置Skill一键回滚执行Skill优化建议输出Skill数据库表映射开发Skill性能隐患识别Skill流水线配置Skill流水线故障排查SkillPRD撰写SkillPrompt工程优化SkillPrompt注入测试Skill质量趋势分析SkillRAG召回效果评测SkillRAG全链路搭建Skill重构方案推荐Skill回归用例筛选Skill发布方案设计Skill整改建议输出Skill需求变更评估Skill需求优先级排序Skill需求调研分析Skill资源用量评估Skill风险等级划分Skillroadmap-planningroot-cause-analysissast-scanscaffold-setupsecurity-audit-reportsensitive-data-identificationtech-debt-quantificationtest-case-generationtest-report-analysistrace-analysisui-automationunit-test-generationuser-story-splittingvector-db-tuningversion-planningwireframe-design
Subagents (31)
| AI应用工程化Sub-Agent | 负责RAG全链路搭建、Prompt工程优化、智能Agent编排、向量库部署调优。当项目涉及检索增强生成、业务Prompt模板设计、多智能体编排、向量数据库选型与调优时使用。 |
| 架构设计Sub-Agent | 负责项目技术架构选型、AI应用架构设计、数据库建模、接口协议定义。当新项目/新模块起步,需要输出技术选型方案、架构文档、ER图、前后端接口规范时使用。 |
| 后端开发Sub-Agent | 负责后端多语言业务代码开发、框架脚手架搭建、数据库表映射开发、大模型服务封装。当需要实现服务端业务逻辑(Java/Python/Go/Node等)、搭建项目脚手架、开发数据访问层或封装LLM调用服务时使用。 |
| 前端开发Sub-Agent | 负责多端页面生成(Web/移动端)、组件库封装、前端工程化配置、AI交互组件开发。当需要实现前端页面、封装通用组件、配置前端构建工程、开发AI对话/可视化等交互组件时使用。 |
| 全栈&AI应用开发主Agent | 全栈&AI应用开发主Agent,统筹全栈前后端、AI工程化、中间件基础设施的研发全流程。当总控(主对话)接到功能开发、技术实现、架构设计、AI能力集成等研发类需求,需要拆解任务并调度下属Sub-Agent(架构设计/后端/前端/AI工程化/中间件)时使用。 |
| 中间件与基础设施Sub-Agent | 负责Redis缓存策略设计实现、消息队列规划开发、API网关配置。当项目需要缓存方案落地、MQ消息架构开发、网关路由鉴权配置时使用。 |
| CI/CD流水线Sub-Agent | 负责CI/CD流水线配置搭建、容器镜像构建打包、多环境配置管理、流水线故障排查。当需要搭建或改造持续集成/交付流水线、构建镜像、管理多环境配置时使用。 |
| 资源与成本管理Sub-Agent | 负责服务器/算力资源用量评估、成本优化方案输出、云资源弹性伸缩配置。当需要评估资源用量、制定降本方案、配置弹性伸缩策略时使用。 |
| 发布与部署Sub-Agent | 负责发布方案设计(灰度/全量)、K8s容器编排部署、环境初始化、一键回滚执行。当需要制定发布方案、执行容器编排部署、初始化环境、处理发布回滚时使用。 |
| 运维与交付主Agent | 运维与交付主Agent,统筹CI/CD流水线、发布部署、线上监控排障、资源成本管控的一级调度者。当测试通过的稳定版本需要流水线构建、发布上线、环境初始化、监控告警、故障排查、成本优化时使用。 |
| 监控与故障排查Sub-Agent | 负责监控指标配置、日志分析定位、分布式链路追踪、线上故障根因分析。当需要配置监控大盘、分析日志定位问题、追踪调用链路、定位线上故障根因时使用。 |
| 需求变更与优先级Sub-Agent | 负责需求变更评估、需求优先级排序、路线图规划。当出现需求冲突或变更请求,需要评估影响面、用RICE/Kano等方法排序需求优先级、制定产品路线图时使用。 |
| 产品与需求主Agent | 产品与需求主Agent,研发链路最前端的一级调度者。负责需求调研拆解、PRD管控、原型交互统筹、需求变更与优先级管理。当总控(主对话)接到产品规划、需求分析、PRD撰写、原型设计、需求排期等产品类任务,需要拆解并调度下属Sub-Agent时使用。 |
| PRD与产品设计Sub-Agent | 负责PRD撰写、信息架构设计、功能规格说明、版本规划。当用户故事与验收标准明确后,需要输出正式产品需求文档、梳理信息架构或制定版本迭代计划时使用。 |
| 原型与交互Sub-Agent | 负责线框图设计、交互流程设计、Figma设计稿生成、设计走查。当PRD明确后需要输出页面线框图、交互流程图,或在Figma中生成/落地设计稿时使用。 |
| 需求分析Sub-Agent | 负责需求调研分析、用户故事拆解、验收标准定义、竞品分析。当需求模糊需要调研澄清、把业务诉求拆解为用户故事并定义可测试的验收标准时使用。 |
| 架构评审Sub-Agent | 负责架构合规性校验、依赖关系分析、设计模式评估、技术债务量化。当有架构设计方案产出,需要复核合规性、分析组件依赖、评估设计模式合理性时使用。架构评审不通过禁止进入代码开发阶段。 |
| 代码审查Sub-Agent | 负责逐行校验编码规范、排查代码逻辑缺陷、识别性能隐患、专项审查AI相关代码安全性。当有代码产出需要审查质量与安全问题,输出问题清单与整改建议时使用。只做审查判定,不修改业务代码。 |
| 质量管控与代码审查主Agent | 质量管控与代码审查主Agent,独立把控代码质量、架构合规性、技术债务治理的一级调度者。当研发产出需要全维度评审、代码质量度量、技术债务治理,或开发→测试的门禁评审时使用。 |
| 代码质量度量Sub-Agent | 负责计算代码圈复杂度、统计单元测试覆盖率、检测重复代码、生成质量趋势报表。当需要用数据量化判定代码质量是否达标时使用。 |
| 技术债务管理Sub-Agent | 负责梳理全项目技术债务、按风险分级排序、输出分批次重构优化方案、跟进债务迭代闭环。当项目存在历史遗留问题,需要债务治理规划时使用。 |
| 代码安全审计Sub-Agent | 负责静态安全测试(SAST)、第三方组件漏洞扫描、硬编码密钥检测、AI代码安全审计。当研发阶段需要扫描代码安全漏洞、依赖组件漏洞、密钥泄露问题时使用。研发阶段同步介入前置防控。 |
| 合规与风险评估Sub-Agent | 负责对标行业法规评估合规差距、划分安全风险等级、输出闭环整改建议、定期生成全局安全审计报告。当需要合规评估、风险分级、整改方案或定期安全审计报告时使用。 |
| 数据安全与隐私Sub-Agent | 负责敏感数据识别、数据脱敏方案制定、传输与存储加密策略落地、数据泄露风险评估。当业务涉及敏感数据,需要识别敏感字段、制定脱敏规则、落地加密策略时使用。贯穿研发、传输、运维全环节。 |
| 基础设施安全Sub-Agent | 负责服务器/容器安全基线检测、容器镜像漏洞扫描、防火墙网络策略配置、权限最小化校验。当部署前需要校验基础设施安全基线、加固网络与权限时使用。运维部署前强制校验。 |
| 安全与合规主Agent | 安全与合规主Agent,全流程贯穿校验的一级调度者(非末端环节)。对开发、测试、运维所有环节做安全审计、基础设施基线管控、数据隐私保护、合规风险评估。当任一环节需要安全审计、基线校验、敏感数据治理、合规评估,或发现疑似安全风险需要介入时使用。 |
| AI应用专项测试Sub-Agent | 负责AI大模型应用的回答准确性评测、幻觉检测、Prompt注入测试、RAG召回效果评测。当系统包含大模型能力,需要专项评测回答质量、安全性与检索效果时使用。 |
| 自动化测试Sub-Agent | 负责单元测试代码生成、接口自动化脚本开发、UI自动化脚本开发、测试报告分析。当需要编写自动化测试脚本、执行自动化任务、分析测试报告时使用。 |
| 功能测试Sub-Agent | 负责基于业务需求生成全量功能测试用例、执行业务场景化测试、分级上报缺陷、筛选回归测试用例。当新功能/需求变更需要功能验证与缺陷管理时使用。 |
| 测试工程主Agent | 测试工程主Agent,统筹功能测试、自动化测试、性能压测、AI专项测试的一级调度者。当评审通过的研发产物需要测试验收、缺陷管理、回归验证时使用。 |
| 性能压测Sub-Agent | 负责压测场景设计、压测脚本编写执行、性能瓶颈分析、调优建议输出。当上线前或大流量场景需要验证系统容量、定位性能瓶颈时使用。 |
Similar rigs
lexosi/IA-Workflow-Template
Reusable Claude Code workflow template — agents, git flow, and session memory for any project.
Minimalist 991 tok ·
szarkans/multi
Use [Multi]ple AI agents for code-review, verifying and questions
Pragmatist 545 tok ·
alirezarezvani/claude-code-tresor
A world-class collection of Claude Code utilities: autonomous skills, expert agents, slash commands, and prompts that supercharge your development workflow
Orchestrator 6.4k tok ·
Noorkhalel/saas-ai-skills
Open-source AI Skills library for building, improving, and scaling modern SaaS applications.
Skill Collector 1.2k tok ·