Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/w1ne/kernelcad-web/kernelcad-from-referencenpx skills add w1ne/kernelCAD-web --skill kernelcad-from-referencegit clone --depth 1 https://github.com/w1ne/kernelCAD-webWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00051 | $0.01578 |
| Opus 5 | $0.00026 | $0.00789 |
| Sonnet 5 | $0.00010 | $0.00316 |
| Haiku 4.5 | $0.00005 | $0.00158 |
Grade A, and why
kernelcad-from-reference scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kernelCAD — from reference (orchestrator)
Reproduce a real-world artifact from one or more reference photographs using kernelCAD primitives. This is an orchestrator skill that names sub-skills and their order. Read this first; then load each sub-skill in sequence as you work through the stages.
Decision tree (1 screen — start here)
Have a written spec with numeric dimensions? → just read kernelcad-authoring
+ use-the-available-kernel,
build single-pass, score.
Simple front-on consumer electronics or passive enclosures with a photo and a
known dimension? → read `photo-to-device/SKILL.md` before blockout. Its
hosted-vs-local reference-asset rule controls generic
`referenceImage()` advice for this path; it is not a mesh
concept or mechanism workflow.
Have a reference photo only? ──┬─ Extract numeric dimensions FIRST
│ (measure visually OR if STL available
│ use trimesh to extract bbox+landmarks),
│ then proceed as if you had a spec.
│
└─ Don't iterate against the 2D-photo scorer —
R5/R16/R18 empirical: it's gameable. Use
the geometric scorer (scripts/scoreMeshVsMesh.ts)
if an STL reference is shipped.
Building an organic-curve outline (brow, grip, sneaker)?
→ use path().smoothSpline() — chained sagittaArc hits solver cliffs.
Building an acetate-bevel-style chamfer on post-cut topology?
→ just call .chamfer(d). The kernel auto-skips edges shorter than 2×d
and emits a clean warning. Do NOT "skip and document."
Iteration mode: visual > scored > spec+photo (R1-R6 empirical).
Single-pass with a detailed spec OUTPERFORMS visual or scored iteration.
Adding the photo to a good spec REGRESSES (R3). Closed-loop scorer-only
iteration plateaus FAR below single-pass (R2 / R16). Only iterate when
you have a CLEAR signal pointing at a SPECIFIC defect.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 113 lines · 51 tokens per session scan A 372a760aa57a
kernelcad-from-reference is a skill published in the GitHub repository w1ne/kernelCAD-web (17 stars, last pushed 4d ago), licensed MIT. It adds 51 tokens to every session and 1,578 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cad
Use when doing parametric 3D modeling, 2D drafting, geometric kernel development, BIM/IFC processing, PCB design, or AutoCAD .NET development. Index of 19 skills: FreeCAD, OpenSCAD, OCCT, CadQuery, KiCad, SolveSpace, QCAD, xBIM, Clipper2, TongWen and more.
aieng-cad-authoring
Active MCP-first CAD authoring skill for AIENG Workbench. Use when the user asks to create, design, draft, model, or iteratively build CAD geometry through the active cad. MCP tools. Do not use for solver/result review, schema/library implementation, or editing legacy IR plans.
apollo-onboarding
Plans and configures a Universal Audio Apollo rig end to end through the uad-console-mcp MCP server, covering the interview questions to ask, which physical input suits which source, the clock master decision for ADAT expanders and other digital gear, ADAT sample-rate and channel-count limits, multi-Apollo Thunderbolt…
cadquery-modeling
Create, edit, or debug parametric 3D CAD parts in CadQuery (Python) that export to STEP for SolidWorks. Use when the request involves CadQuery, .step/.stl export, or modelling a physical part — enclosure, bracket, PCB, motor mount, cam, gear. Also use when a STEP will not open in CAD, or geometry looks correct in STL…
sdf
SDFormat/SDF model and world authoring, validation, and simulator handoff. Use for .sdf files, SDFormat XML, models, worlds, links, joints, poses, frames, inertials, visual/collision geometry, mesh URIs, sensors, lights, physics, plugins, includes, Gazebo, static SDF review, or simulator-specific metadata. Do not use…
urdf
URDF robot description authoring and validation. Use when creating, editing, inspecting, validating, or debugging .urdf files, robot links, joints, limits, inertials, visual/collision geometry, mesh references, frame conventions, or robot-description artifacts. Use the SRDF skill for MoveIt2 semantic groups and…