Skip to content

Strategic Guide - Metrics ​

This is the single source of truth for the numbers that describe KernDX: how much code it contains, how many tests run, how big the public API is, and how it compares on size and activity. Every other guide links here instead of repeating a figure, so there is only one place to keep current. If you are looking up a statistic to cite, this is the page. Update this file when the codebase changes.

Part of the KernDX Strategic Guide.

Last verified: 2026-11-17 (counts recomputed from the source tree for the 1.12 release, including 80 Jest test files and 3,597 Jest test cases, each test declaration counted once; package-version rows re-verified the same day against sfdx-project.json at 1.12.0-3) Source: Counts derive from the force-app/ source tree and from git log on the Last verified date above. Update this file when the codebase changes.


Table of Contents ​

Expand
  1. Package Codebase
  2. Org Apex Usage (Unmanaged)
  3. Global API Surface
  4. Extension Points (Where Your Code Connects)
  5. LWC Components
  6. Salesforce Metadata
  7. Documentation
  8. Code Quality & Scanning
  9. Release Testing
  10. E2E Tests (Playwright)
  11. Combined Test Coverage
  12. Activity Snapshot
  13. The Other Frameworks We Compared

Package Codebase ​

MetricCountNotes
Apex classes (total)410sum of production + test
Apex production classes204force-app/main/default/classes/*.cls excluding _TEST.cls
Apex test classes206force-app/main/default/classes/*_TEST.cls
Apex test methods6,380counted from force-app/main/default/classes/*_TEST.cls (method-level @IsTest annotations; class-level @IsTest(...) annotations are excluded)
Apex test coverage100% per-fileenforced at every release build by scripts/evaluate-coverage.js
Lines of code (production Apex)~105Kcounted from force-app/main/default/classes/*.cls excluding _TEST.cls
Lines of code (test Apex)~114Kcounted from force-app/main/default/classes/*_TEST.cls
Lines of code (production LWC)~27Kcounted from force-app/main/default/lwc/**/*.js excluding .test.js
Lines of code (LWC Jest tests)~49Kcounted from force-app/main/default/lwc/**/*.test.js
Lines of code (total)~294Kthe rows above are rounded; exact total 294,036
Source API version68.0visible in Setup → Installed Packages → Kern → API Version; also in sfdx-project.json
Latest packaged version1.12.0-3 (released)sfdx-project.json packageAliases

Org Apex Usage (Unmanaged) ​

MetricValue
Apex characters in use1,209,561
Apex character limit10,000,000
Percent of limit used12.10%

The org measured here is the development org, which is a scratch org, and Salesforce gives scratch orgs a 10 MB limit (Apex governor limits: "For scratch orgs, the limit is 10MB."). Against the standard 6 MB limit, the same code would use about 20%.

You can see this figure yourself in Setup → System Overview. The count includes every Apex class and trigger defined in the org, excluding comments and test classes (those annotated @IsTest). Managed package code does not count towards this limit, which is why installing KernDX as a managed package leaves your org's own Apex headroom untouched.

Global API Surface ​

MetricCount
Global classes (top-level)109
Global interfaces (top-level)4
Global inner classes~94
Global inner interfaces~26
Global inner enums~16
Global methods~556

Extension Points (Where Your Code Connects) ​

The points where your own code connects to the framework. The Global API Surface section above counts the raw symbols (classes, methods, and so on). This section is more useful day to day: it groups those connection points by how you use them, so you can tell at a glance whether a given extension needs configuration, Apex, or just clicks in Flow.

Metadata-Configured (no code required) ​

MetricRecordsNotes
TriggerSetting__mdt7
TriggerAction__mdt11
FeatureFlag__mdt15excluding the TestFeatureFlag fixture record
FeatureFlagStrategy__mdt6excluding the TestFeatureFlagStrategy fixture record
ApiSetting__mdt4
ApiCredential__mdt3
MaskingRule__mdt183 active by default (MaskSecretKeys, MaskPaymentCard, MaskJwt), 15 shipped inactive (14 templates and the legacy MaskCreditCard that MaskPaymentCard replaces)
MaskingTarget__mdt14Wildcards wiring the card and secret rules to ApiCall__c, ApiIssue__c, AsyncChainExecution__c, and LogEntryEvent__e, and the JWT rule to ApiCall__c and ApiIssue__c (ApiCall_All_Jwt, Issue_All_Jwt); 4 of the 14 belong to the replaced credit-card rule and ship inactive
ApiMock__mdt0Extensibility-only: no pre-built records, you add your own
ValidationRuleGroup__mdt0Extensibility-only: no pre-built records, you add your own
ValidationRule__mdt0Extensibility-only: no pre-built records, you add your own
ClassTypeResolver__mdt0Extensibility-only: no pre-built records, you add your own
AsynchronousJobSetting__mdt1JobRunnerCheckLimit, for the scheduled job runner; you add your own for other jobs
FieldSetGroup__mdt0Extensibility-only: no pre-built records, you add your own
PostTriggerAction__mdt0Extensibility-only: no pre-built records, you add your own
Total pre-built custom metadata (CMDT) records79excluding the two Test* fixture records

Code-Level (extend or implement) ​

MetricCountComposition
Base classes to extend8SEL_Base, TRG_Base, API_Outbound, API_Inbound, SCHED_Base, DTO_JsonBase, UTIL_TypeResolver.BaseClassResolver, TST_Builder.DefaultValueProvider
Top-level interfaces4IF_Queryable, IF_Schedulable, IF_Search, IF_TableDataSource
IF_Trigger inner interfaces7BeforeInsert, AfterInsert, BeforeUpdate, AfterUpdate, BeforeDelete, AfterDelete, AfterUndelete
IF_Async inner interfaces2Processable, Finishable
Other inner interfaces4IF_Chain.Step, QRY_Condition.Evaluable, UTIL_TypeResolver.INT_ClassTypeResolver, SVC_Omnistudio.OmniCallable
ComponentBuilder LWC modules6notification, controller, navigation, lightning-message, flow-navigation, all

Declarative (Flow Builder / Lightning App Builder) ​

MetricCount
Flow invocable actions (FLOW_*)16
Exposed LWC components33
LWC with Lightning page targets15
— App Page11
— Home Page7
— Record Page6
— Flow Screen5
— Tab1
— Record Action1
Pre-built flows6
Platform events1 (LogEntryEvent__e)

LWC Components ​

MetricCountNotes
LWC components (total)78force-app/main/default/lwc/**/
Jest test files80counted from force-app/main/default/lwc/**/*.test.js
Jest test cases3,597each test declaration counted once; a Jest run reports more, because it runs a table-driven test once per row
Jest test coverage95%+ statement / 95%+ branchenforced at every release build by scripts/evaluate-coverage.js

Salesforce Metadata ​

MetricCount
Custom objects (__c)11
Custom metadata types (__mdt)15
Platform events (__e)1
Custom fields (total)304
— on custom objects162
— on custom metadata types122
— on platform events20
Pre-built CMDT records79

Documentation ​

MetricCountNotes
Developer guides21Excluding the Strategic Guide series; counted from docs/*%20-%20Guide.md
Fast Start guides17
Strategic Guide documents9Overview + Architecture & Philosophy + Adoption + Operations + Risks + Glossary + Personas + Metrics + Choosing a Framework
API reference pages (Apex)252
API reference pages (metadata)16
API reference pages (objects)12
API reference pages (events)2
API reference pages (total)282sum of all reference categories
Total documentation files (developer-focused)4721 developer guides + 17 Fast Start guides + 9 Strategic Guide documents (excluding reference/)
Security Guide (lines)2,511bespoke threat-model guide; counted from docs/Security - Guide.md

The headline figure quoted in companion docs is 38 developer documents (21 guides + 17 Fast Starts), matching the per-category totals above.

Code Quality & Scanning ​

MetricCount
PMD rules (KernDX custom)28
ESLint rules (KernDX custom)8
Node validators4
Total scanner checks40

Release Testing ​

These tests run in a separate org with the package installed the way you would install it (the rows call it the subscriber org: the org a package is installed into). The end-to-end (E2E) browser tests are counted under E2E Tests (Playwright) below.

MetricCount
Subscriber Apex classes (production)88
Subscriber test classes66
Subscriber test methods388
Anonymous Apex test sections85
Anonymous Apex assertions (PASS:)463
Anonymous Apex assertions (Assert.)595

Local test run ​

The package's own Apex tests, run on a computer with no Salesforce org. The figures below are written by scripts/aer-run.js into verify-results/aer-run.json, and the docs check fails when this table and that file disagree.

MetricValue
Apex tests passing locally6,370 of 6,380 (99.84%)
Runtimeaer 1.6.10 by October Swimmer, a local Apex runtime
How it ranas a user with the Administrator permission set (-p Administrator), on a copy of the source with no profiles folder, in one run
Sourcev1.12.0-3
Measured2026-11-18
Failing: aer differs from Salesforce10
Failing: cause not yet known0

What the runtime cannot run: the two "Failing" rows above. aer copies most of how Salesforce behaves, not all of it, so a real org stays the final word. Each failing test is listed with its reason in scripts/aer-baseline.json. On this run 10 tests failed. Each one passes in a Salesforce org, and each one fails on aer because aer behaves differently from Salesforce:

Where aer differs from SalesforceTest it affects
aer counts the Apex type System.Location as a record object, so finding the Location object by name fails with a type error; Salesforce does notQRY_BuilderBypassSharing_TEST.bypassSharingAloneChecksATypeofWhoseWhenNamesLocationLikeAnyObject
aer finds no event for a filter on the phone of the person the event is about (Who.Phone); Salesforce finds the eventQRY_BuilderBypassSharing_TEST.bypassSharingAloneReadsANameRecordFieldThatTheRelatedObjectHides
aer finds no event for a filter on the last name of the person the event is about (Who.LastName); Salesforce finds the eventQRY_BuilderBypassSharing_TEST.bypassSharingAloneReadsAPathThatCanPointToAnObjectTheUserCannotRead
aer looks for What.ProfileId on Account and the query fails; Salesforce reads it from Name, the shared object that stands for any record What can point toQRY_BuilderBypassSharing_TEST.bypassSharingAloneReadsTheUserFieldsOfTheNameRecordForAStandardUser
In user mode (the user's own permissions apply), aer runs a query that sorts on a parent field the user cannot read; Salesforce refuses itQRY_BuilderBypassSharing_TEST.bypassSharingAloneRefusesAHiddenFieldOfAParentTheUserCanRead
In user mode, aer runs a query that selects a field through a lookup the user cannot read; Salesforce refuses itQRY_BuilderBypassSharing_TEST.bypassSharingAloneRefusesAPathThroughALookupTheUserCannotRead
In user mode, aer runs a query that selects What.Name when the user cannot read WhatId; Salesforce refuses itQRY_BuilderBypassSharing_TEST.bypassSharingAloneRefusesASelectedPathThroughALookupTheUserCannotRead
In user mode, aer runs a TYPEOF query that names an object the user cannot read; Salesforce refuses itQRY_BuilderBypassSharing_TEST.bypassSharingAloneRefusesATypeofWithATypeTheUserCannotRead
For a Minimum Access user, aer says RecordTypeId on Name can be read, so the query fails; Salesforce says it cannot, so KernDX refuses the query before it runsQRY_BuilderBypassSharing_TEST.bypassSharingAloneRefusesTheUserFieldsOfTheNameRecordForAMinimumAccessUser
aer counts the Apex type System.Location as a record object, so finding the Location object by name fails with a type error; Salesforce does notUTIL_SObjectDescribe_TEST.shouldFindObjectWhoseNameIsAlsoSystemTypeName

E2E Tests (Playwright) ​

MetricCount
Spec files15
Test cases132
Page objects14
Helper modules8
CMDT fixture states24

The Test cases row counts the functional release-verification specs (parts 1 to 12). Three further spec files sit outside that count: two capture the documentation screenshots (15 cases between them) and one checks the Log Console's shipped screens against the agreed design (3 cases). The Spec files row counts all 15 files.

Combined Test Coverage ​

LayerTestsAssertions
Package Apex tests6,380 methods~7,729
Package Jest tests3,597 cases~5,054
Subscriber Apex tests388 methods~349
Subscriber anonymous Apex85 sections463
Subscriber E2E (Playwright)132 cases~342
Total10,582~13,937

The Tests column is recounted from source every release. The Assertions column carries forward from the 1.6 reconciliation and was not recounted this release, except the anonymous Apex row, which takes the PASS: count from the Release Testing table above.

Activity Snapshot ​

These figures describe how well-supported and ready-to-install the package is, measured on 2026-11-17. They are kept separate from the capability counts on purpose: how long a project has been active should not be confused with how much it does, so neither number skews the other. Figures derive from sfdx-project.json packageAliases.

MetricValueSource
Published managed-package versions151sfdx-project.json packageAliases: 04tfj000000EtB7AAK through 04tfj000000lFlBAAU
Latest packaged version1.12.0-3 (released)sfdx-project.json packageAliases
Primary contributors1single-maintainer project

KernDX is a single-maintainer project. Its readiness to install shows up in the 151 published managed-package versions, rather than in a long public commit history. So what this means for you: the package itself is well-exercised through repeated releases, even though it does not yet have the years-long open commit log of an older project.

The Other Frameworks We Compared ​

When the guides rank or compare KernDX, they measure it against the same group of other Salesforce frameworks across the same capability areas, so every comparison is like-for-like. This page only points you to the detail rather than repeating it: the Adoption and Architecture guides walk through how each area is covered, and the Risks guide covers maintenance and the bus-factor question (what happens if the people who built it move on).


Strategic Guide (Overview)