How to add a kernel

This is the extension recipe contributors should follow.

Checklist

Step 0 – Decide scope

Support: positive-only vs real line
Tail option: bulk-only vs bulk+GPD
Backend compatibility: CRP, SB, or both

Step 1 – Implement kernel functions

Implement the kernel’s d/p/q/r surface (and any internal helpers). Keep vectorization and support checks consistent.

Step 2 – Implement Mix / MixGPD wrappers

If the kernel is used inside mixture components or spliced tails, implement the corresponding MixGPD functions.

Step 3 – Register it

Add an entry to the kernel registry so compile_model_spec() can discover it.

Step 4 – Tests

Add at least: - support checks (values outside support return 0 density, etc.) - monotonic CDF / quantile inversion checks - a tiny end-to-end bundle build test

Step 5 – Docs

Add website/kernels/kernel-<name>.qmd
Add link from website/kernels/kernels-index.qmd
Mention the kernel in kernel_support_table() documentation

Templates

R code skeleton: developers/templates/kernel-skeleton.R
Registry entry skeleton: developers/templates/registry-entry-skeleton.R
Test skeleton: developers/templates/test-kernel-skeleton.R

Prereqs

Required packages and data for this page are listed in the setup chunks above.

Outputs

This page renders model fits, diagnostics, and summary artifacts generated by package APIs.

Interpretation

Canonical concept page: Spec And Contracts
Treat this page as an application/example view and use the canonical page for core definitions.

Continue to the linked canonical concept page, then return for implementation-specific details.