Use this skill when configuring a new or existing Marketing Cloud Account Engagement (MCAE, formerly Pardot) business unit: Salesforce v2 connector activation, connector user permissions, Salesforce User Sync, prospect field sync rules, tracking domain setup, and Account Engagement Lightning App installation. NOT for Marketing Cloud Engagement (email studio, journeys) or Einstein Marketing features.

RecordType ID differences across orgs: dynamic lookup via Schema.SObjectType methods, DeveloperName for stable refs, hard-coded ID anti-patterns, record-type caching, deployment impact. NOT for record-type strategy or when to use record types (use admin-foundation or data-model-design).

Use this skill when eliciting, documenting, and structuring requirements for a Salesforce implementation or enhancement: writing Salesforce-specific user stories with acceptance criteria, mapping As-Is and To-Be business processes, conducting stakeholder discovery interviews, and performing gap analysis against standard Salesforce capabilities. Trigger keywords: requirements gathering, user story, As-Is To-Be, gap analysis, stakeholder interview, process mapping, business requirements, fit gap. NOT for technical design decisions (use solution-design-patterns). NOT for automation implementation (use flow/* or apex/* skills). NOT for data model design (use data-model-design-patterns or object-creation-and-design).

Revenue Intelligence setup: pipeline inspection, deal insights, forecast accuracy analytics, Einstein analytics for sales leaders. NOT for CRM Analytics platform admin (use analytics-studio-admin). NOT for forecasting category setup only (use forecasting-and-quotas).

Use this skill when gathering or evaluating requirements for a Salesforce Enterprise Territory Management (ETM) territory design: alignment criteria, coverage model selection, assignment rule logic, geographic considerations, hierarchy depth and breadth, and user-to-territory ratios. Trigger keywords: territory design, territory alignment, territory model requirements, sales coverage model, territory criteria, geographic territory, named account territory, overlay territory. NOT for ETM configuration or setup steps — use enterprise-territory-management for that. NOT for role hierarchy design — use sharing-and-visibility.

Use this skill when writing acceptance criteria for Salesforce features, structuring UAT test scripts from user stories, classifying defects found during UAT, or planning regression testing before a Salesforce release. Trigger keywords: UAT, user acceptance testing, test script, acceptance criteria, defect classification, regression testing, test plan, test case. NOT for automated testing (use flow-testing or apex-test-class-standards). NOT for writing user stories or eliciting requirements (use requirements-gathering-for-sf). NOT for setting up sandboxes or environment strategy (use sandbox-strategy).

Apex test patterns for @InvocableMethod agent actions: per-branch coverage, bulk safety, deterministic assertions. NOT for UI/LWC testing or agent conversational quality scoring.

Canonical go-live checklist for Agentforce deployments with rehearsed rollback and stakeholder sign-off records. NOT for general Salesforce release management (see release-management).

Authoring and managing Agentforce agent definitions using the declarative Agent Script DSL (.agent files) and associated metadata types. Use when creating agents in source control, debugging agent metadata, or understanding the metadata lifecycle of GenAiPlugin/GenAiPlanner/BotVersion types. Triggers: 'how do I deploy an Agentforce agent using source control', 'what metadata types make up an Agentforce agent', 'agent test run command failing in CI pipeline', 'GenAiPlugin vs GenAiPlanner metadata relationship'. NOT for Apex-based agent actions (use custom-agent-actions-apex). NOT for UI-based agent creation (use agentforce-agent-creation).

Use when testing, evaluating, or building regression suites for Agentforce agents: conversation testing in Agent Builder, topic coverage and utterance testing, Testing API and AiEvaluationDefinition metadata, evaluation metrics (containment rate, escalation rate, CSAT, topic activation accuracy), and post-deploy analytics via Enhanced Event Logs. Triggers: 'how do I test my Agentforce agent', 'agent routes to wrong topic', 'write utterance tests', 'regression test after topic change', 'measure agent quality', 'agent containment rate'. NOT for agent creation, topic design, or action contract design — use agentforce/agentforce-agent-creation, agentforce/agent-topic-design, or agentforce/agent-actions respectively.

Use when creating, configuring, auditing, or troubleshooting an Agentforce agent end-to-end: agent definition, agent user setup, channel assignment, system instructions, activation, and lifecycle management. Triggers: 'create agentforce agent', 'agent not appearing to users', 'how to activate agent', 'agent channel setup', 'agent lifecycle', 'deploy agent to production'. NOT for topic design or action contract design — use agentforce/agent-topic-design and agentforce/agent-actions respectively.

Author and run offline evals for Agentforce agents: fixture format, scoring rubrics, regression baselines, CI integration, prompt-change safety. Use BEFORE every prompt or tool change. Covers multi-turn transcripts, refusal checks, tool-call correctness, grounding accuracy. NOT for online A/B testing (use observability). NOT for general Salesforce test-class patterns (use apex-testing-patterns).

Use when configuring Slack-specific Agentforce capabilities AFTER the basic Slack deployment is complete: enabling Slack-native actions, managing public vs. private action scope, mapping Salesforce-to-Slack user identity, or troubleshooting Slack action failures. Triggers: 'add General Slack Actions topic to agent', 'configure canvas creation for Agentforce in Slack', 'Agentforce private action requires user identity mapping', 'Slack agent cannot send DMs after deployment', 'Agentforce Look Up User action not working in Slack', 'how to map Slack users to Salesforce identities for private actions'. NOT for core Agentforce setup, NOT for basic Slack OAuth installation or DM vs channel-mention mode — those are covered by agentforce/agent-channel-deployment.

Design Agentforce conversations that span multiple turns without losing context: session variable scoping, conversation memory, clarifying-question patterns, topic-to-topic handoff, and the right abstractions for accumulating state across turns. NOT for single-turn agent actions (use agent-actions). NOT for channel-specific conversation UX (use agent-channel-deployment).

Use when monitoring Agentforce agent sessions, analyzing conversation logs, measuring deflection rates, or diagnosing agent performance issues. Triggers: 'agentforce session analytics', 'how to query agent conversation data', 'monitor agentforce agent effectiveness', 'agent deflection rate', 'utterance analysis agentforce'. NOT for Einstein Trust Layer audit logging (use einstein-trust-layer), NOT for agent topic design or guardrails (use agent-topic-design or agentforce-guardrails), NOT for LLM prompt debugging (this skill covers session metrics and conversation trace, not prompt engineering).

Step-by-step setup and configuration of Einstein for Sales AI features: Opportunity Scoring, Pipeline Inspection AI insights, Einstein email insights and composition, and Forecasting AI. Covers prerequisites, license checks, feature sequencing, and data readiness validation. NOT for core Agentforce agent creation, agent topic design, Einstein Trust Layer configuration, or Einstein Activity Capture troubleshooting.

Design Agentforce testing: topic coverage, action unit tests, deterministic golden sets, adversarial prompts, and regression harness. Trigger keywords: agentforce testing, agent eval, agent regression suite, prompt golden set, action unit test agentforce. Does NOT cover: generic LLM evaluation academia, human-labeled RLHF pipelines, or Einstein Classify accuracy.

Pick the right tool shape for each agent action: Apex invocable vs Flow action vs External Service vs Prompt Template vs Data Cloud retrieval. Covers action selection by use case, argument design for LLM clarity, return-shape contracts, error-surfacing, cost implications, and when to chain tools vs keep a single action. NOT for authoring a specific action (use custom-agent-actions-apex). NOT for topic design (use agent-topic-design).

Lifecycle management for Prompt Builder templates: version, test, promote, roll back via CMDT-backed bindings. NOT for authoring initial templates or generic prompt engineering.

Use when structuring Apex into service, selector, domain, factory, and dependency-injection layers for maintainability and testability. Triggers: 'service layer', 'selector pattern', 'domain layer', 'dependency injection', 'fat trigger/controller'. NOT for installing a specific third-party framework or debating package-level architecture outside Apex code structure.

Use when choosing between DML statements and Database class methods, handling partial success, managing savepoints, or using Database.DMLOptions for assignment rules and duplicate handling. Trigger keywords: 'Database.insert allOrNone false', 'partial DML success apex', 'SaveResult isSuccess', 'database merge apex', 'DML exception handling'. NOT for SOQL query patterns (use soql-fundamentals or apex-soql-relationship-queries), NOT for sharing model setup (use apex-managed-sharing-patterns).

Execute Anonymous Apex: Developer Console, VS Code SFDX, `sf apex run`, variable scope, transaction semantics, governor limits, debugging output, common errors. NOT for scheduled/queueable async (use apex-async-patterns). NOT for debug log settings (use debug-log-management).

HttpCalloutMock for Apex tests: HttpCalloutMock interface, StaticResourceCalloutMock, MultiStaticResourceCalloutMock, Test.setMock, multi-call mocks for pagination, error-path mocks. NOT for the callout code itself (use callouts-and-http-integrations). NOT for WSDL callouts (use apex-wsdl2apex-patterns).

Use when serializing Apex objects to JSON strings or deserializing JSON responses into Apex types — especially for callout payloads, integration parsing, and controlling null field output. Trigger keywords: 'suppress null fields JSON Apex', 'deserialize JSON into Apex class', 'JSON parse unknown shape', 'TypeException JSON deserialize', 'JSONGenerator streaming'. NOT for REST endpoint response shaping (use apex-rest-services), NOT for Apex remote actions returning JSON to LWC.

Use when choosing or implementing Apex test doubles with `Test.setMock`, `HttpCalloutMock`, `StaticResourceCalloutMock`, or `StubProvider`, and when designing code seams to support those doubles cleanly. Triggers: 'StubProvider', 'Test.createStub', 'HttpCalloutMock', 'StaticResourceCalloutMock', 'mocking infrastructure'. NOT for general Apex test design outside the mocking and seam problem.

Use when diagnosing where Apex transactions spend CPU, heap, SOQL, or DML time using the Salesforce diagnostic toolchain: Apex Log Analyzer flame graphs in VS Code, Developer Console execution timeline, SOQL Query Plan tool, and Limits-class checkpoint instrumentation. Triggers: 'why is my Apex slow', 'flame graph Apex', 'profile Apex transaction', 'Query Plan tool', 'Apex Log Analyzer'. NOT for fixing specific CPU or heap patterns after the hotspot is found (use apex-cpu-and-heap-optimization), NOT for debug log setup or trace flag mechanics (use debug-logs-and-developer-console), and NOT for SOQL query rewriting (use soql-query-optimization).

System.runAs in Apex tests: user-context impersonation, mixed-DML workaround, profile/permission testing, sharing verification, FLS NOT enforced, runAs nesting limits. NOT for general test setup (use apex-test-setup-patterns). NOT for WITH USER_MODE SOQL (use apex-user-mode-patterns).

Use when writing Apex triggers on the Case object — specifically for invoking assignment rules programmatically, auto-associating entitlements in a trigger, handling merge trigger behavior on losing records, or understanding why milestone completion does not fire automatically when a case closes. Trigger keywords: 'case trigger', 'case assignment rule apex', 'entitlement auto-association trigger', 'case merge trigger', 'MasterRecordId case', 'Database.DmlOptions AssignmentRuleHeader', 'milestone not completing on case close'. NOT for generic trigger framework architecture — use apex/trigger-framework for that. NOT for configuring assignment rules in Setup — use admin/assignment-rules. NOT for SLA configuration or entitlement process design — use admin/entitlements-and-milestones.

Use when diagnosing or preventing performance problems caused by cross-object formula spanning relationships and roll-up summary field recalculations — especially in LDV orgs. Triggers: 'roll-up summary timing stale in trigger', 'Maximum 15 object references error', 'rollup recalculation timeout on large child set'. NOT for formula syntax authoring (use admin/formula-fields), formula compile-size limits (use apex/formula-field-performance-and-limits), or general governor-limit troubleshooting.

Use when setting up debug logs and trace flags, reading Apex log output and log levels, running queries in the Developer Console, executing anonymous Apex, or using the Apex Replay Debugger in VS Code. Triggers: 'set up debug log', 'Developer Console', 'anonymous Apex', 'trace flag', 'Apex Replay Debugger'. NOT for production logging strategy or custom structured logging frameworks (use debug-and-logging), and NOT for writing test classes (use apex-test-class-standards).

Programmatic record access management in Financial Services Cloud using Compliant Data Sharing (CDS): inserting AccountParticipant/OpportunityParticipant records, working with ParticipantRole and ParticipantGroup objects, verifying sharing coverage, and understanding CDS recalculation behavior. NOT for admin setup of CDS via IndustriesSettings metadata, declarative sharing rules, or standard Apex managed sharing on non-FSC objects.

Use this skill when implementing Apex code that calls the FSL scheduling Apex API: GetSlots, schedule(), GradeSlotsService, or OAAS optimization classes. Trigger keywords: AppointmentBookingService, ScheduleService.schedule, GetSlots, GradeSlotsService, OAAS, bulk scheduling, FSL optimization callout. NOT for admin-level scheduling policy configuration, scheduling rules UI setup, or the standard Book Appointment quick action without custom Apex.

Use this skill when building custom headless storefronts that call Salesforce Commerce API (SCAPI) — including Shopper API authentication with SLAS OAuth2 JWT tokens, integrating Commerce SDK React hooks, wiring Shopper Basket/Checkout APIs, and handling SCAPI-specific load-shedding and caching behavior. NOT for standard LWR storefront configuration, declarative Experience Builder layouts, legacy OCAPI integrations, or architecture-level headless decisions — use admin/headless-commerce-architecture for pattern selection.

Use this skill when extending Health Cloud via Apex: implementing HealthCloudGA managed-package interfaces, automating care plan lifecycle hooks, processing referrals using Industries Common Components invocable actions, or enforcing HIPAA-compliant logging governance for clinical Apex code. Trigger keywords: CarePlanProcessorCallback, HealthCloudGA namespace, ReferralRequest, ReferralResponse, care plan invocable actions, clinical Apex extension, Health Cloud Apex API. NOT for standard Apex triggers or generic Apex development unrelated to Health Cloud managed-package extension points.

Use this skill when working with Health Cloud APIs: querying healthcare-specific SObjects (CarePlan, ClinicalEncounter, HealthCondition) via standard SObject API, using the FHIR R4-aligned Healthcare API, handling FHIR bundle limits, and understanding API authentication differences between the two layers. NOT for generic REST API development or standard Salesforce SObject API patterns unrelated to Health Cloud clinical objects.

Use this skill when integrating with the Account Engagement (Pardot) REST API v5 to create, read, update, or query Prospects, Visitors, Visitor Activities, or Form Handlers programmatically. Trigger keywords: Pardot API, Account Engagement API, prospect sync, visitor tracking API, form handler integration, MCAE REST API. NOT for Marketing Cloud API (Email Studio, Journey Builder REST), NOT for standard Salesforce CRM REST API, NOT for MCAE setup or business unit provisioning.

Use when encountering or preventing MIXED_DML_OPERATION errors caused by DML on setup objects (User, UserRole, PermissionSet, Group, GroupMember) in the same transaction as non-setup objects. Triggers: 'MIXED_DML_OPERATION', 'setup object DML error', 'cannot insert User and Account together', 'System.runAs mixed DML', '@future setup object workaround'. NOT for general async Apex patterns — see async-apex. NOT for test data factory structure — see test-data-factory-patterns.

Use when extending the NPSP Trigger-Driven Trigger Management (TDTM) framework with custom Apex handler classes — covering class authorship, DmlWrapper return patterns, Trigger_Handler__c registration, load order, recursion guards, and test isolation. NOT for standard Apex triggers outside of NPSP, general trigger-handler framework design, or Nonprofit Cloud (NPC) which replaced NPSP in new orgs.

Use this skill to implement Apex-driven custom routing logic for Omni-Channel work items using PendingServiceRouting and SkillRequirement objects. Trigger keywords: PendingServiceRouting, SkillRequirement, IsReadyForRouting, skills-based routing, custom routing Apex. NOT for declarative Omni-Channel queue-based routing setup, routing configurations in Setup UI, or Einstein Classification routing rules.

Use when debugging or preventing recursive Apex trigger behavior, especially around self-DML, static guard flaws, Set<Id>-based deduplication, and legitimate re-entry scenarios. Triggers: 'trigger recursion', 'static boolean guard', 'recursive update', 'self DML', 'trigger firing multiple times'. NOT for general trigger-framework structure unless recursion is the actual design problem.

Use when enrolling records in Sales Engagement cadences from Apex, logging call outcomes on cadence steps, customizing cadence step actions via Apex, or consuming cadence lifecycle events through Change Data Capture. Triggers: 'enroll lead in cadence', 'assignTargetToSalesCadence', 'ActionCadenceTracker', 'log call result cadence', 'Sales Engagement Apex'. NOT for building or administering cadence structures (steps, branches, variants) — cadence content must be authored in the Cadence Builder UI and cannot be created or mutated via API.

Use this skill when integrating with Salesforce Service Cloud REST APIs — specifically the Knowledge REST API (/knowledgeManagement/ for authoring, /support/knowledgeWithSEO/ for guest retrieval) and the Messaging for In-App and Web (MIAW) Enhanced Chat API. Trigger keywords: Knowledge article retrieval, data category filtering, URL-name lookup, Enhanced Chat API, legacy Chat REST API migration. NOT for generic REST API callouts from Apex (use apex__http-callouts), NOT for Salesforce REST API CRUD on standard objects, NOT for Einstein Bot configuration.

Use when designing or building reusable Apex test data factories: @IsTest utility classes, SObject hierarchy construction, bulk data generation, portal user factories with System.runAs(), and @testSetup methods. NOT for test class structure and assertions (use apex/test-class-standards) or for loading data into sandboxes (use devops/data-seeding-for-testing).

Use this skill when designing or evaluating the holistic Salesforce AI platform strategy: model tier selection (Salesforce Default vs BYOLLM vs zero-data-retention partner models), Trust Layer design decisions at the architecture level, and multi-agent Supervisor/Specialist orchestration topology. Trigger keywords: AI platform strategy, model selection strategy, Supervisor agent topology, multi-agent orchestration, Einstein LLM gateway, BYOLLM architecture, trust layer design, agentic platform design, model tier decision. NOT for individual agent action development, Trust Layer feature configuration steps, RAG grounding mechanics, or BYOLLM registration procedures (see agentforce/einstein-trust-layer, agentforce/rag-patterns-in-salesforce, agentforce/model-builder-and-byollm for those).

Use Parchi Relay to control a real browser via the Parchi extension agent (WebSocket) or directly via JSON-RPC from a CLI/tooling workflow.

>

This document defines the patterns and workflows for working with the shared-gha repository.

>-

Create, refactor, evaluate, and package agent skills from workflows, prompts, transcripts, docs, or notes. Use when asked to create a skill, turn a repeated process into a reusable skill, improve an existing skill, add evals, or package a skill for team reuse.

Comprehensive operational guide for agents using Google Ads CLI.