agentleFS
wedsamuel1230wedsamuel1230/embedded-engineering-skills/skills/embedded-build-workflow/SKILL.md

embedded-build-workflow

Skill28 starsChanged 29 days agoMITView on GitHub

Inspect, reproduce or migrate embedded build workflows with PlatformIO, native ESP-IDF, Arduino CLI, Zephyr/west, Pico SDK, STM32Cube or MicroPython. Use when a request concerns build environments, mixed frameworks, dependency pins, CI, SDK migration or flash/PSRAM configuration. Do not use for ordinary application logic, GPIO lookup or flashing an unknown device.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the shared contract. Preserve source locations and existing pinned tooling before proposing migration.

    line 14
  2. Very specific:

    Run the offline inspector: uv run scripts/inspect_build.py --project PATH.

    line 16
  3. Distinctive:

    Read toolchain branches for the detected framework.

    line 21
  4. Very specific:

    If PlatformIO contains extra_configs, extends or interpolation, obtain its native computed configuration and retain unresolved status …

    line 23
  5. Distinctive:

    Do not share raw computed config when it contains credentials.

    line 25
  6. Very specific:

    Keep migration staged: baseline build, bounded API/driver adaptation, host tests, native build, then authorized target tests.

    line 34
  7. Distinctive:

    Use the selected specialist for concrete native mechanics: Zephyr, STM32Cube, Pico SDK, MicroPython.

    line 70

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Embedded Build Workflow

Produce an explicit project/environment matrix and the smallest reproducible build path. A framework name is not target support or a successful build.

Workflow

  1. Read the shared contract. Preserve source locations and existing pinned tooling before proposing migration.
  2. Run the offline inspector: uv run scripts/inspect_build.py --project PATH. It never builds, flashes, opens ports, installs packages or executes project scripts. It returns argument arrays, not shell commands. Treat custom build scripts as code to review before a separately authorized build.
  3. Read toolchain branches for the detected framework. Choose the repository's build manager, not an Arduino default. If PlatformIO contains extra_configs, extends or interpolation, obtain its native computed configuration and retain unresolved status meanwhile. Do not share raw computed config when it contains credentials.
  4. Record board/carrier revision, module suffix, flash and PSRAM separately, environment, host OS, tool versions and dependency resolution. For ESP32-S3 memory work, read memory evidence; use scripts/check_memory_config.py only with the explicit physical profile and provenance of the selected generated sdkconfig.
  5. Build every declared program/environment in scope. If an input program has no build path, report it; do not infer that every sketch needs a sibling INI when a root project already owns its source. Reuse the same matrix in CI.
  6. Keep migration staged: baseline build, bounded API/driver adaptation, host tests, native build, then authorized target tests. Arduino libraries may rely on Wire, timing, callbacks and runtime initialization unavailable natively.
  7. Report the inspected evidence and unverified stages. Use the support matrix, not a family-wide badge.

Outputs

  • project paths; all declared environments; framework and manager;
  • explicit versions, unresolved inheritance and custom-script trust boundary;
  • build/test argument arrays and actual results, when execution was authorized;
  • four memory outcomes: confirmed-mismatch, consistent-with-inspected-evidence, unresolved, not-applicable;
  • source revision/input hashes, selected environment and artifact identity;
  • upload, physical memory population and runtime behavior marked unverified.

Ownership

The workflow router retains combined board/wiring/application work. arduino-cli-skill owns Arduino CLI commands; native projects stay here. board-support resolves board identity; this skill checks build configuration. realtime-control-design, freertos-patterns and wireless-telemetry own their respective designs. Keep pin and electrical claims outside the inspector.

Sources and verification

The helpers use the Python standard library. Run uv run -m unittest discover -s tests -p test_embedded_workflows.py from the package root. See source lessons and research lineage. Tests of the helper prove its output and uncertainty handling, not a firmware build. Run the trusted compile fixtures separately before claiming build proof.

Platform procedures

Use the selected specialist for concrete native mechanics: Zephyr, STM32Cube, Pico SDK, MicroPython. Rust is outside the current implementation scope.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.