harnesse2e

qa-engineer 的默认 skill。USE FOR: 测试用例设计、执行验证、测试报告输出、缺陷分级。

unite-test 的默认 skill。USE FOR: 测试用例设计、执行验证、测试报告输出、缺陷分级。

developer 的默认 skill。USE FOR: 按设计实现代码、补齐测试、控制技术债、保证可运行性。

Use for Android middleware, AOSP, Android Automotive, Binder, AIDL, HIDL, HAL, SystemServer, CarService, Service, APK and framework-native boundary architecture from requirements and design documents.

Generate, revise, 补全 or重写 car middleware, cockpit middleware, IVI middleware architecture from PRD,需求文档,requirements/requirements_spec.md or feature specs. Use for Linux/QNX/Android cross-OS 架构设计, 中间件分层, 服务拓扑, 通信矩阵, 资源预算, 诊断, 安全 and design/architecture.md.

Use for C++ middleware, native service, performance critical module, real-time path, memory ownership, threading, IPC bridge, ABI boundary and reliability architecture from automotive or embedded requirements and design docs.

Use for Java architecture in middleware and platform projects: module layering, domain model, service orchestration, concurrency model, exception boundary, interface contract and maintainable package design from requirements and architecture tasks.

Use for Linux middleware, Linux daemon, systemd service, POSIX IPC, D-Bus, Unix socket, shared memory, epoll, watchdog and resource isolation architecture in cockpit, IVI or embedded platform requirements and architecture docs.

Use for QNX Neutrino middleware, PPS, resource manager, channel/message passing, Screen, io-pkt, adaptive partitioning, high reliability and real-time service architecture in cockpit, cluster or embedded requirements and architecture docs.

Generate, revise, 补全, 重写 or收口 architecture documents from PRD,需求文档,需求规格说明书,requirements/requirements_spec.md or feature specs. Use for 输出 design/architecture.md, and when requested also design/api_design.md and design/data_model.md, with API 边界, 模块划分, 数据流, 状态流, 技术选型, 安全设计 and 可观测性设计 for backend, platform, Android, IVI and cockpit projects.

用于根据功能规范、功能说明、产品草案、原始需求输入生成软件需求文档。适用于需求拆解、表格化功能需求编写、验证准则整理、数据与配置约束梳理、待确认项整理。

用于检查软件需求文档质量门禁。适用于需求评审、需求质量检查、需求表完整性校验、验证准则检查、待确认项和工程化补足合规检查、输出放行判定。

Default programming backend for this agent. Delegate coding/editing/refactor/build/test tasks to the GitHub Copilot CLI (`copilot` command) via bash. Use for: writing new features, fixing bugs, refactoring, running tests, exploring an unfamiliar codebase. Copilot is the preferred backend; fall back to Claude Code only if Copilot fails. Never run inside ~/.openclaw, /Users/clawbot/.npm-global/lib/node_modules/openclaw, or any OpenClaw state directory.

build-verifier 的默认 skill。USE FOR: 构建验证、CI/CD 结果汇总、质量门禁收口、发布前放行判定。

code-reviewer 的默认 skill。USE FOR: 代码评审、问题分级、风险判断、只审不改的评估输出。

debugger 的默认 skill。USE FOR: 定位失败原因、最小范围修复、回归验证建议、失败后补救。

delivery-manager 的默认 skill。USE FOR: 交付收口、门禁确认、提交建议、外部通知摘要生成。

planner 的默认 skill。USE FOR: 任务拆解、阶段规划、依赖梳理、质量门禁设计。

safety-reviewer 的默认 skill。USE FOR: 车载功能安全评审、标准条款映射、ASIL 与合规风险检查。

security-reviewer 的默认 skill。USE FOR: 权限与注入风险审计、敏感信息检查、生产风险识别。

基于 ADB 的 Android/车机实车测试技能。USE FOR: 根据需求文档和测试用例执行真机/实车测试，采集 adb shell、logcat、dumpsys、截图与录屏证据，输出结构化测试报告。关键词: adb、实车测试、车机测试、真机测试、Android Automotive、测试报告、requirements、test cases。

节省 token 消耗的最佳实践。USE FOR: 优化 prompt 长度、减少重复输出、控制上下文大小、提高成本效益。