salesforce-apex-quality
Apex code quality guardrails for Salesforce development. Enforces bulk-safety rules (no SOQL/DML in loops), sharing model requirements, CRUD/FLS security, SOQL injection prevention, PNB test coverage (Positive / Negative / Bulk), and modern Apex idioms. Use this skill when reviewing or generating Apex classes, trigger handlers, batch jobs, or test classes to catch governor limit risks, security gaps, and quality issues before deployment.
- License 许可
- MIT
- 最近更新
- 2026-11-13
- Stars
- ★ 4.1w
- Forks
- 5.4k
01 前言 Front Matter
- name
- salesforce-apex-quality
- description
- Apex code quality guardrails for Salesforce development. Enforces bulk-safety rules (no SOQL/DML in loops), sharing model requirements, CRUD/FLS security, SOQL injection prevention, PNB test coverage (Positive / Negative / Bulk), and modern Apex idioms. Use this skill when reviewing or generating Apex classes, trigger handlers, batch jobs, or test classes to catch governor limit risks, security gaps, and quality issues before deployment.
02 资源文件 3
- 参考文档agents/openai.yaml341 B
- 参考文档references/RESEARCH-AND-MAPPING.md7.6 KB
- 参考文档references/VERIFICATION-AND-CUTOVER.md4.4 KB
03 SKILL.md 正文
Salesforce Apex Quality Guardrails
Apply these checks to every Apex class, trigger, and test file you write or review.
Step 1 — Governor Limit Safety Check
Scan for these patterns before declaring any Apex file acceptable:
SOQL and DML in Loops — Automatic Fail
// ❌ NEVER — causes LimitException at scale
for (Account a : accounts) {
List<Contact> contacts = [SELECT Id FROM Contact WHERE AccountId = :a.Id]; // SOQL in loop
update a; // DML in loop
}
// ✅ ALWAYS — collect, then query/update once
Set<Id> accountIds = new Map<Id, Account>(accounts).keySet();
Map<Id, List<Contact>> contactsByAccount = new Map<Id, List<Contact>>();
for (Contact c : [SELECT Id, AccountId FROM Contact WHERE AccountId IN :accountIds]) {
if (!contactsByAccount.containsKey(c.AccountId)) {
contactsByAccount.put(c.AccountId, new List<Contact>());
}
contactsByAccount.get(c.AccountId).add(c);
}
update accounts; // DML once, outside the loop
Rule: if you see [SELECT or Database.query, insert, update, delete, upsert, merge inside a for loop body — stop and refactor before proceeding.
Step 2 — Sharing Model Verification
Every class must declare its sharing intent explicitly. Undeclared sharing inherits from the caller — unpredictable behaviour.
| Declaration | When to use |
|---|---|
public with sharing class Foo |
Default for all service, handler, selector, and controller classes |
public without sharing class Foo |
Only when the class must run elevated (e.g. system-level logging, trigger bypass). Requires a code comment explaining why. |
public inherited sharing class Foo |
Framework entry points that should respect the caller's sharing context |
If a class does not have one of these three declarations, add it before writing anything else.
Step 3 — CRUD / FLS Enforcement
Apex code that reads or writes records on behalf of a user must verify object and field access. The platform does not enforce FLS or CRUD automatically in Apex.
// Check before querying a field
if (!Schema.sObjectType.Contact.fields.Email.isAccessible()) {
throw new System.NoAccessException();
}
// Or use WITH USER_MODE in SOQL (API 56.0+)
List<Contact> contacts = [SELECT Id, Email FROM Contact WHERE AccountId = :accId WITH USER_MODE];
// Or use Database.query with AccessLevel
List<Contact> contacts = Database.query('SELECT Id, Email FROM Contact', AccessLevel.USER_MODE);
Rule: any Apex method callable from a UI component, REST endpoint, or @InvocableMethod must enforce CRUD/FLS. Internal service methods called only from trusted contexts may use with sharing instead.
Step 4 — SOQL Injection Prevention
// ❌ NEVER — concatenates user input into SOQL string
String soql = 'SELECT Id FROM Account WHERE Name = \'' + userInput + '\'';
// ✅ ALWAYS — bind variable
String soql = [SELECT Id FROM Account WHERE Name = :userInput];
// ✅ For dynamic SOQL with user-controlled field names — validate against a whitelist
Set<String> allowedFields = new Set<String>{'Name', 'Industry', 'AnnualRevenue'};
if (!allowedFields.contains(userInput)) {
throw new IllegalArgumentException('Field not permitted: ' + userInput);
}
Step 5 — Modern Apex Idioms
Prefer current language features (API 62.0 / Winter '25+):
| Old pattern | Modern replacement |
|---|---|
if (obj != null) { x = obj.Field__c; } |
x = obj?.Field__c; |
x = (y != null) ? y : defaultVal; |
x = y ?? defaultVal; |
System.assertEquals(expected, actual) |
Assert.areEqual(expected, actual) |
System.assert(condition) |
Assert.isTrue(condition) |
[SELECT ... WHERE ...] with no sharing context |
[SELECT ... WHERE ... WITH USER_MODE] |
Step 6 — PNB Test Coverage Checklist
Every feature must be tested across all three paths. Missing any one of these is a quality failure:
Positive Path
- Expected input → expected output.
- Assert the exact field values, record counts, or return values — not just that no exception was thrown.
Negative Path
- Invalid input, null values, empty collections, and error conditions.
- Assert that exceptions are thrown with the correct type and message.
- Assert that no records were mutated when the operation should have failed cleanly.
Bulk Path
- Insert/update/delete 200–251 records in a single test transaction.
- Assert that all records processed correctly — no partial failures from governor limits.
- Use
Test.startTest()/Test.stopTest()to isolate governor limit counters for async work.
Test Class Rules
@isTest(SeeAllData=false) // Required — no exceptions without a documented reason
private class AccountServiceTest {
@TestSetup
static void makeData() {
// Create all test data here — use a factory if one exists in the project
}
@isTest
static void givenValidInput_whenProcessAccounts_thenFieldsUpdated() {
// Positive path
List<Account> accounts = [SELECT Id FROM Account LIMIT 10];
Test.startTest();
AccountService.processAccounts(accounts);
Test.stopTest();
// Assert meaningful outcomes — not just no exception
List<Account> updated = [SELECT Status__c FROM Account WHERE Id IN :accounts];
Assert.areEqual('Processed', updated[0].Status__c, 'Status should be Processed');
}
}
Step 7 — Trigger Architecture Checklist
- [ ] One trigger per object. If a second trigger exists, consolidate into the handler.
- [ ] Trigger body contains only: context checks, handler invocation, and routing logic.
- [ ] No business logic, SOQL, or DML directly in the trigger body.
- [ ] If a trigger framework (Trigger Actions Framework, ff-apex-common, custom base class) is already in use — extend it. Do not create a parallel pattern.
- [ ] Handler class is
with sharingunless the trigger requires elevated access.
Quick Reference — Hardcoded Anti-Patterns Summary
| Pattern | Action |
|---|---|
SOQL inside for loop |
Refactor: query before the loop, operate on collections |
DML inside for loop |
Refactor: collect mutations, DML once after the loop |
| Class missing sharing declaration | Add with sharing (or document why without sharing) |
escape="false" on user data (VF) |
Remove — auto-escaping enforces XSS prevention |
Empty catch block |
Add logging and appropriate re-throw or error handling |
| String-concatenated SOQL with user input | Replace with bind variable or whitelist validation |
| Test with no assertion | Add a meaningful Assert.* call |
System.assert / System.assertEquals style |
Upgrade to Assert.isTrue / Assert.areEqual |
Hardcoded record ID ('001...') |
Replace with queried or inserted test record ID |
04 同仓库更多 Skills
supabase-rls-policies
Write, review, and test Supabase Row Level Security (RLS) policies so that one user cannot read or change another user's rows. Use when creating or changing tables, policies, views, or security definer functions in a Supabase project, when reviewing a migration that touches RLS, or when asked to prove user isolation with pgTAP tests that sign in as two users.
skill-distribution-audit
Audit a skill collection before release or investigate installer omissions, archived/template leakage, duplicate identities, and manifest drift. Reconcile the intended public source set, distribution manifest, and actual discovery output for a specified consumer version. Includes a deterministic set-comparison helper.
logicapps-custom-code-net10-migration
Migrate Azure Logic Apps Standard local-function custom-code projects from .NET 8 to .NET 10. Use for legacy Logic Apps custom-code workspace upgrades, project/package configuration, custom-worker settings, publication, and integrated VS Code debugging. Not for ordinary .NET applications, standalone Azure Functions, Consumption workflows, or cloud deployment.
winmd-api-search
Find and explore Windows desktop APIs. Use when building features that need platform capabilities — camera, file access, notifications, UI controls, AI/ML, sensors, networking, etc. Discovers the right API for a task and retrieves full type details (methods, properties, events, enumeration values).
resemble-detect
Deepfake detection and media safety — detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity, and analyze media intelligence using Resemble AI
quality-playbook
Run a complete quality engineering audit on any codebase. Derives behavioral requirements from the code, generates spec-traced functional tests, runs a three-pass code review with regression tests, executes a multi-model spec audit (Council of Three), and produces a consolidated bug report with TDD-verified patches. Finds the 35% of real defects that structural code review alone cannot catch. Works with any language. Trigger on 'quality playbook', 'spec audit', 'Council of Three', 'fitness-to-purpose', or 'coverage theater'.