analyze制品一致性分析技能。对 spec.md、design.md、tasks.md 做跨制品一致性分析并自动修复,不设轮数上限,全程不询问用户。残留问题完整打印后继续正常结束。用于分析制品、检查一致性、质量审查。
archive将变更指示文件中的需求/设计/接口变更回流到 DOC_DIR/specs 下的要素文档. 当需要从 feature 变更目录同步更新到 DOC_DIR/specs 下的需求、场景、实体、功能和接口文档及其清单与关联关系时使用
brainstorming-sdd-bridgeExpert workflow adapter that converts an approved brainstorming *-design.md into the standard SDD document interface consumed by tasks/implement/review/local-sandbox-fix. Use only after expert brainstorming and before tasks.
brainstormingYou MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
checklist检查清单生成技能. 基于用户需求为当前功能生成自定义检查清单, 清单是"需求编写的单元测试", 验证需求的质量、清晰度和完整性而非实现. 当用户提到生成检查清单、需求质量验证、规范完整性检查时使用.
clarify规范澄清技能. 通过提出最多 5 个高度针对性的澄清问题, 识别当前功能规范中未充分说明的领域, 并将答案编码回规范中. 当用户提到澄清规范、消除模糊性、补充规范细节、或在 specify 完成后需要进一步细化时使用.
code-reviewProactively reviews code for quality, security, and maintainability. Use immediately after writing or modifying code. MUST BE USED for all code changes.
cpp-coding-standardsC++ coding standards based on the C++ Core Guidelines (isocpp.github.io). Use when writing, reviewing, or refactoring C++ code to enforce modern, safe, and idiomatic practices.
cpp-testingUse only when writing/updating/fixing C++ tests, configuring GoogleTest/CTest, diagnosing failing or flaky tests, or adding coverage/sanitizers.
create-branch为功能开发准备工作环境——创建或复用 Git 分支与功能目录(changes/...),并初始化 spec.md 规格文档。当需要开始新功能开发、修复缺陷、创建分支、新建功能、切换分支、或切换到特定功能上下文时使用。
design-entity基于已定义的功能识别并调整逻辑实体,输出实体责任边界、协作关系与关键数据结构,并生成数据模型文档
design-function从场景出发,分析业务意图对既有功能的影响,生成变更的功能,并追加到 IMPL_DESIGN 末尾
design-interface基于逻辑实体的能力边界生成接口契约,输出对外接口与内部协作接口的契约要素与变更规则,并生成接口契约文档
design执行实施规划工作流, 使用计划模板生成设计制品,在代码之前建立可执行的技术决策体系。
django-securityDjango security best practices, authentication, authorization, CSRF protection, SQL injection prevention, XSS prevention, and secure deployment configurations.
e2e-design测试实现分析:在黑盒测试用例基础上,分析入口函数、外部依赖(Fake)、测试数据、验证点,复用存量测试设计。当用户说"/e2e-design"、"测试实现分析"、"生成e2e-impl-design"、"测试设计细化",或需要基于测试用例和设计文档生成测试实现分析报告时使用此技能。
e2e-specify测试分析与设计(TA&TD):使用MFQ&PPDCS方法论进行测试分析,从需求提取测试点(M单功能/F功能交互/Q质量属性),然后使用TCON方法设计黑盒测试用例。输出:测试分析报告(test-analysis.md)和黑盒测试用例(e2e-test.md)。当用户说"/e2e-specify"、"生成测试设计"、"创建测试用例"、"分析测试设计",或在规范生成后需要生成测试设计时使用此技能。
e2e-varifyE2E测试设计完整性检查与完善。对测试实现分析报告进行深度检查,包括变更点覆盖分析、外部依赖异常场景、测试数据设计完善。当用户说"/e2e-varify"、"完善测试设计"、"验证测试用例"、"测试用例审查",或需要将测试设计草稿打造成完善的可指导测试代码落地的文档时使用此技能。
eval-code-evaluatorEvaluate code generation quality using third-party judge model metrics (ICE Score + Code Judge). Use this when the user needs to assess generated code quality, compare code outputs, or evaluate AI-generated code against requirements. This skill provides automated scoring for functional correctness,
eval-code代码质量综合评测:先采集SDD流程的代码变更信息,再使用第三方评测模型对代码质量进行评估。当用户需要"综合评测"、"完整评测流程"、"SDD评测"时触发本技能。
eval-design-consistency-check需求一致性检查知识库,提供需求边界一致性检查、Scope Creep 检测、需求覆盖检测、语义一致性分析三类维度及代码读取判断业务归属策略。由 requirement-consistency-reviewer(Robin)引用,适用于 spec_review、design_scope_check、detail_design_review 三种审核模式。
eval-design对 design.md 进行多维度质量评测,输出 badcase 清单与 surface_reason。通过阈值:score >= 95 可进入下一环节。
eval-specifySpec 产出物内容质量评测量规。定义四维评分(业务价值、技术完整性、清晰度与可测性、可追溯性)的检查项、打分标准、YAML 报告格式。由 evaluation-analyst(Eva)引用,适用于 spec_eval 模式。
fastapi-patternsFastAPI patterns for async APIs, dependency injection, Pydantic request and response models, OpenAPI docs, tests, security, and production readiness.
golang-patternsIdiomatic Go patterns, best practices, and conventions for building robust, efficient, and maintainable Go applications.
golang-testingGo testing patterns including table-driven tests, subtests, benchmarks, fuzzing, and test coverage. Follows TDD methodology with idiomatic Go practices.
implement执行实施计划,通过处理并执行 tasks.md 中定义的所有任务完成实施。强制遵循 TDD(RED→GREEN→REFACTOR)范式。
incremental-coverage仅针对增量(新增/修改)文件收集代码覆盖率。适用于功能分支覆盖率验证、TDD 第 7 步、 implement 技能完成后、PR 需要覆盖率证据等场景。支持 C/C++、Java、Python、Go、 JavaScript/TypeScript、Rust, 通过语言无关的流水线实现: 识别变更文件 -> 开启覆盖率插桩 -> 运行目标测试 -> 提取过滤覆盖率 -> 生成可操作报告。
java-coding-standardsJava coding standards for Spring Boot and Quarkus services: naming, immutability, Optional usage, streams, exceptions, generics, CDI, reactive patterns, and project layout. Automatically applies framework-specific conventions.
jpa-patternsJPA/Hibernate patterns for entity design, relationships, query optimization, transactions, auditing, indexing, pagination, and pooling in Spring Boot.
knowledge-retrieval私域项目知识检索与知识库构建。支持两种模式:enhance(用户提供知识模型,typed 精细检索)与 baseline(无建模要求,chunk 窗口切片向量 + graphify 图谱兜底)。当需要从项目知识沉淀中获取信息以辅助需求分析、方案设计、编码、测试,或需要构建/刷新知识索引时使用。
local-sandbox-fix基于 local-sandboxcheck 检查结果自动修复代码并重测。提交前本地 CI 闭环。触发词:local-sandbox-fix、本地沙盒修复、sandbox 自动修复。
mini-design-review评审极简SDD生成的详设文档
mini-design极简SDD生成详设流程,根据输入的方案要求结合代码生成详设文档。由 `/mini-design` 调用本 skill(与技能名同名)。
mini-implement-review评审极简SDD生成的代码
mini-implement极简SDD根据详设生成代码流程。由 `/mini-implement` 调用本 skill(与技能名同名)。
python-patternsPythonic idioms, PEP 8 standards, type hints, and best practices for building robust, efficient, and maintainable Python applications.
python-testingPython testing strategies using pytest, TDD methodology, fixtures, mocking, parametrization, and coverage requirements.
quarkus-patternsQuarkus 3.x LTS architecture patterns with Camel for messaging, RESTful API design, CDI services, data access with Panache, and async processing. Use for Java Quarkus backend work with event-driven architectures.
quarkus-securityQuarkus Security best practices for authentication, authorization, JWT/OIDC, RBAC, input validation, CSRF, secrets management, and dependency security.
quarkus-tddTest-driven development for Quarkus 3.x LTS using JUnit 5, Mockito, REST Assured, Camel testing, and JaCoCo. Use when adding features, fixing bugs, or refactoring event-driven services.
report-persistencePersists review results (security or code review) to structured JSON and Markdown files for traceability. Called as a sub-step after completing any review.
reverse-on-demand按需反构编排Skill。基于需求意图执行代码架构分析,支持简单/复杂需求场景。当 reverse --target on-demand 时触发。
routing智能路由编排器. 分析功能描述复杂度, 将 flow_mode(express/standard/deep/expert)映射到对应 YAML 工作流定义并启动 workflow-orchestrator skill 执行。触发词: /routing, 智能编排, 路由模式.
runlog-record记录 omni skill 执行后的维测数据到 JSON 文件。当需要统一记录 skill 执行结果时、当 omni 工具执行完成后需要生成执行报告时、当需要进行 skill 性能分析时自动触发。适用于 SDD workflow 各阶段结束后手动调用。
rust-patternsIdiomatic Rust patterns, ownership, error handling, traits, concurrency, and best practices for building safe, performant applications.
rust-testingRust testing patterns including unit tests, integration tests, async testing, property-based testing, mocking, and coverage. Follows TDD methodology.
sddOmniSpec 全流程开发入口(SDD 范式)。用户调用 /sdd 或说"部署omnispec"时触发,执行环境初始化后调用 routing skill,由 routing 决定 workflow 模式并启动对应 agent。
security-reviewUse this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
skill-craftEvaluates, repairs, creates, and audits Claude Skills using an 8-module quality framework with weighted scoring. Performs structural scanning, anti-pattern detection, completeness verification, discovery validation, and decision-gate checks. Use when checking Skill quality ("check skill", "skill 质量"
skill-validator验证Claude Skill的规范性、完整性和最佳实践。检查SKILL.md文件的前置元数据、内容结构、目录组织和代码规范,支持检查、交互式修复和自动化修复三种模式。适用于skill开发、审查、优化和质量保证场景。
spec-impact-analyze需求波及分析:从需求出发分析已有规格与代码知识,判断可复用范围与波及影响,产出结构化上下文供 context.md 使用,不直接写文件。
specify-requirement从系统视角分析业务意图对既有需求的影响,生成变更需求(INSERT/MODIFY/DELETE/REFER),并追加到 spec.md。仅被 /specify skill 显式调用,不自动触发。使用 EARS 需求表达法进行结构化需求描述。
specify-scenario分析业务意图对既有场景的影响,基于 INSERT/MODIFY/DELETE/REFER 动作类型生成变更场景,并将场景追加到 spec.md 中。适用场景:用户提到"场景变更"、"新增场景"、"修改场景"、"Given/When/Then"、"验收用例"等关键词,或需要更新功能需求文档中的场景章节时。输出包含场景ID、优先级、归属需求、验收用例等结构化内容。仅被 specify skill 显式调用,不自动触发。
specify从自然语言功能描述生成功能规范文档(SDD)。当用户输入 /specify 或提供功能需求描述时自动触发。执行上下文收集、需求分析、场景提取,生成规范文件。适用于需要从需求生成规范,建立功能说明等场景。触发关键词:/specify、规范生成、功能描述、spec.md、SDD。
springboot-patternsSpring Boot architecture patterns, REST API design, layered services, data access, caching, async processing, and logging. Use for Java Spring Boot backend work.
springboot-securitySpring Security best practices for authn/authz, validation, CSRF, secrets, headers, rate limiting, and dependency security in Java Spring Boot services.
springboot-tddTest-driven development for Spring Boot using JUnit 5, Mockito, MockMvc, Testcontainers, and JaCoCo. Use when adding features, fixing bugs, or refactoring.
springboot-verificationVerification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.
tasks基于可用的设计文档为功能特性生成可执行的、按依赖关系排序的 tasks.md。在用户要求生成任务列表、制定实施计划、从 spec/design 产出 tasks 时使用。
tdd-workflowUse this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 90% line / 70% branch coverage including unit, integration, and E2E tests.
workflow-orchestrator统一工作流编排器。读取 YAML 工作流定义,按 stage.type 派发:skill 直接 Skill 触发,agent 经 Task(general-purpose) 隔离执行。
writing-tasksUse when you have a spec or requirements for a multi-step task, before touching code