workflow
当需要串联多个 skill 完成编译+烧录+监控或编译+烧录+调试等流水线任务时使用。
Install
npx skills add https://github.com/leokemp223/embed-ai-tool --skill workflowSKILL.md
Workflow 流水线编排
适用场景
- 需要一键完成编译、烧录、串口监控的完整开发流程。
- 需要编译后自动烧录并启动 GDB 调试。
- 需要在不同构建系统(Keil/CMake/PlatformIO)间使用统一的流水线接口。
必要输入
- 构建系统类型(keil、cmake、platformio)。
- 工程路径。
- 可选:构建目标、串口、波特率、烧录参数。
依赖
- 对应构建系统的 skill 脚本已存在(build-keil、flash-keil 等)。
- 各 skill 的外部依赖已安装(Keil UV4、CMake、PlatformIO CLI 等)。
执行步骤
- 先阅读 references/usage.md,确认参数。
- 探测环境:
python scripts/workflow_runner.py --detect - 查看可用流水线:
python scripts/workflow_runner.py --list - 执行流水线:
python scripts/workflow_runner.py --run build-flash-monitor --build-system platformio --project /path/to/project
失败分流
environment-missing:对应 skill 脚本不存在。target-response-abnormal:某个步骤执行失败(编译错误、烧录失败等)。
输出约定
示例输出格式:
🚀 执行流水线: build-flash-monitor(编译 → 烧录 → 串口监控)
构建系统: platformio
==================================================
[1/3] 编译
==================================================
$ python .../platformio_builder.py --project-dir /path
==================================================
[2/3] 烧录
==================================================
$ python .../pio_flasher.py --flash --project-dir /path
==================================================
[3/3] 串口监控
==================================================
$ python .../serial_monitor.py --listen --port COM42
📊 结果: ✅ 流水线完成(3 步)
交接关系
- 编排
build-keil/build-cmake/build-platformio的编译步骤。 - 编排
flash-keil/flash-openocd/flash-platformio的烧录步骤。 - 编排
serial-monitor的监控步骤。 - 编排
debug-gdb-openocd/debug-platformio的调试步骤。
Related skills
find-skillsvercel-labs3.6MHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.handoffmattpocock883KCompact the current conversation into a handoff document for another agent to pick up.microsoft-foundrymicrosoft618KBuild, deploy, evaluate, optimize, fine-tune, and manage Microsoft Foundry agents, models, and resources end to end. USE FOR: foundry, azd ai agent, azd provision/deploy, hosted agent scaffold/develop/run/deploy/troubleshoot, prompt agent create, create agent, update agent, add tool to agent, invoke agent, agent.yaml, agent insights, pull agent insights, evaluate agent, batch eval, continuous eval, continuous monitoring, agent CI/CD, optimize prompt, improve prompt, prompt optimizer, optimize agcavemanjuliusbrussee544KUltra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".