Claude Skills by JMMonte

Add or edit a material in a hardware-as-code project's cited material source-of-truth (design/materials.json). Use whenever an analysis needs a material property (density, modulus, Poisson, yield/ultimate strength, composite lamina properties) — instead of letting the model invent one. Enforces: NEVER hallucinate; every property carries a primary-source citation (MMPDS / CMH-17 / ASM / manufacturer datasheet); composites cite the specific prepreg + fiber-volume-fraction + layup, never "generi...

Add a new executable scored check (PASS/WARN/FAIL) for a requirement to a hardware-as-code project's audit harness. Use when a requirement, clearance, mass budget, margin, or anchor needs to be verified automatically. Enforces the rules: read the build's exported JSON artifacts, never duplicate a geometry constant, assert an EXPECTED value (not just ">0"), and exit 1 on FAIL.

Run the continuum (tier-3) FEA on a part a beam model can't capture — bracket, fitting, lug, hull bottom panel, pressure shell, sponson root — with gmsh + CalculiX (ccx). Use when an audit's analytic check or the in-house beam FEA is too coarse and you need a real 3-D stress/displacement or buckling field. Drives the gmsh tet-mesh -> Abaqus .inp -> ccx -> parse -> VALIDATE-against-the-analytic-tier -> rerun pipeline, headless, from a uv single-file script. Refuses to trust a single mesh.

Run the silent-failure guardrails checklist before trusting a sim, audit, build, or tradeoff result in a hardware-as-code project — and record new lessons when something silently gives a wrong answer. Use when a number surprises you, before declaring a result correct, or when reviewing a geometry/sim/audit change. This is the living-checklist entry point: it both APPLIES the gotchas and APPENDS new ones.

Bring an EXISTING CAD file (STEP/.stp, or STL/OBJ/GLB/PLY mesh) into the hardware-as-code pipeline so it can be verified and simulated — the "I already have the CAD" path, as opposed to authoring geometry as code. Use when the user provides a CAD/mesh file and wants audits, FEA, CFD, mass/CG, clearances, or rerun viz on it. Loads the file, converts to canonical mm, computes the artifacts the audits expect (out/imported.json + a mesh), and wires up materials for mass. Handles the units / water...

Drive the automation-agent manufacturability/packaging loop for a hardware-as-code project: an agent authors engineering judgment (decomposition into fabricated vs COTS, placements, harness, sourcing) and automation scores it (skin containment, envelope collisions, mass closure, CG, retrieval corridors), looping until it packs. Use when laying out internals, decomposing a design for fabrication, or packaging COTS into an envelope.

Resolve a hardware design fork with a scored tradeoff study instead of a gut call. Use when a real design decision branches (layout, structure, battery placement, sponsons, strut geometry, ...). Drives a placement-level study scored on computed physics, a sensitivity sweep across weight schemes to pick the ROBUST option, and a tiled side-by-side rerun comparison (one Spatial3DView per candidate). Only the winner earns a full build + sim.

Scaffold a new simulation-in-the-loop, hardware-as-code project at best-practice on day one. Use when starting a new hardware/CAD/sim model, a digital twin, or any project where geometry is code and requirements must be verified by executable audits. Creates the design-of-record (design.json) + a build-source mirror stub, a scored-audit harness, a rerun viz harness, a tradeoff-study template, a manufacturability judgment+solver pair, an AGENTS.md with the repository-map convention and the gua...