Service-specific AI features in Service Cloud: Case Classification setup and optimization, Article Recommendations configuration, Reply Recommendations, Work Summary (After-Visit Summary), Service Replies with Einstein, Auto-Routing, and Einstein Conversation Mining. NOT for core Agentforce setup, agent topic design, Agent Builder configuration, or Einstein Trust Layer setup.

Einstein Prediction Builder for creating custom binary classification predictions on Salesforce objects: field selection, training data requirements, model activation, score embedding, and monitoring. NOT for Einstein Discovery (Tableau CRM Analytics) or Agentforce agent creation.

Use when creating, reviewing, or troubleshooting Prompt Builder templates (Field Generation, Record Summary, Sales Email, or Flex types), including grounding with merge fields, Flow, or Apex. Trigger keywords: prompt template, Prompt Builder, field generation, record summary, sales email template, flex template, grounding, merge fields, LLM template, Einstein generative AI. NOT for agent topic instructions, Copilot action configuration, or Data Cloud segment activation.

Red-team an Agentforce agent against prompt-injection and jailbreak attacks; codify test cases and guardrails. NOT for general application-security reviews outside the agent boundary.

Use this skill when writing, debugging, or reviewing AMPscript in Marketing Cloud email bodies, subject lines, preheaders, SMS, push notifications, or Cloud Pages — including Lookup/LookupRows data retrieval, IF/ELSEIF conditional blocks, FOR loops over rowsets, and inline personalization. NOT for Server-Side JavaScript (SSJS), REST API calls from content, SQL Query Activities, or Journey Builder configuration.

Use when designing, reviewing, or debugging Apex code that relies on List, Set, or Map collections in triggers, batch classes, or service layers — especially for bulkification, heap management, and safe null handling. Trigger keywords: 'Map<Id, SObject>', 'containsKey', 'retainAll', 'putAll', 'Set intersection', 'heap limit', 'collection in loop', 'unbounded accumulation'. NOT for SOQL query optimization (use soql-fundamentals), NOT for async job design (use apex-queueable-patterns or batch-apex-patterns), NOT for platform cache strategies (use platform-cache).

@future methods: primitive-only parameters, callout=true, no chaining, 50 per transaction, error handling. When to prefer Queueable/Batch instead per async-selection decision tree. NOT for Queueable patterns (use apex-queueable-patterns). NOT for Batch Apex (use apex-batch-patterns).

Use when designing, implementing, reviewing, or debugging Queueable Apex jobs that chain, use the Finalizer interface, pass state across transactions, or need controlled async depth. Trigger keywords: 'Queueable', 'System.enqueueJob', 'Finalizer', 'QueueableContext', 'AsyncOptions', 'stack depth', 'chained queueable'. NOT for basic async Apex mechanism selection (use async-apex), NOT for large-volume record processing where Batch Apex is the right tool (use batch-apex-patterns).

Use when building, reviewing, or debugging inbound Apex REST resources, request/response handling, status codes, versioned URL mappings, or JSON serialization in `@RestResource` classes. Triggers: 'Apex REST', '@RestResource', 'HttpGet/HttpPost', 'RestContext', 'versioned endpoint'. NOT for outbound HTTP callouts or standard Salesforce REST API usage as a consumer.

Scheduling Apex classes using the Schedulable interface: implementing execute(), cron expressions, System.schedule(), monitoring CronTrigger records, job limits, and job chaining patterns. NOT for Batch Apex scheduling (use batch-apex-patterns) or Flow scheduled paths.

Use when designing, reviewing, or debugging Apex execution context, sharing keywords, CRUD/FLS enforcement, system-vs-user mode behavior, or secure write patterns. Triggers: 'with sharing', 'inherited sharing', 'stripInaccessible', 'AuraEnabled security', 'CRUD FLS'. NOT for SOQL injection review alone — use apex/soql-security for query-specific hardening.

@TestSetup method semantics: one-time creation per test class, isolation behavior, @TestVisible, System.runAs, Test.startTest/stopTest governor reset, mixed-DML boundaries. NOT for building a test data factory (use test-data-factory-patterns). NOT for mocking callouts (use apex-http-callout-mocking).

Apex Trigger.new / Trigger.old / Trigger.newMap / Trigger.oldMap / Trigger.isInsert etc.: when each is populated, null-safety, recursion depth, trigger event matrix. NOT for trigger framework architecture (use apex-trigger-handler-framework). NOT for bulk patterns (use apex-bulkification-patterns).

Use when selecting, designing, or reviewing Queueable, Batch, Future, or Schedulable Apex for callouts, large data processing, retries, or background work. Triggers: 'queueable vs batch', 'future method', 'flex queue', 'async job failed', 'schedule apex'. NOT for Platform Events or a deep-only Batch Apex implementation guide.

Use when designing, reviewing, or debugging Batch Apex contracts, scope sizing, stateful behavior, chaining, and AsyncApexJob monitoring. Triggers: 'Database.Batchable', 'Database.Stateful', 'executeBatch', 'batch scope', 'AsyncApexJob'. NOT for generic async choice discussions where Queueable or Future might still be the better tool.

Use when building, reviewing, or debugging outbound Apex HTTP callouts, Named Credentials, request/response handling, timeout behavior, or mock-based tests. Triggers: 'HttpRequest', 'Named Credential', 'callout exception', 'uncommitted work pending', 'HttpCalloutMock'. NOT for inbound Apex REST service design or non-HTTP integration architecture.

Use this skill when writing or reviewing Apex change event triggers: trigger syntax on ChangeEvent objects, reading ChangeEventHeader fields (changeType, changedFields, recordIds, commitUser), handling CREATE/UPDATE/DELETE/UNDELETE and GAP events in Apex, and configuring entity tracking. NOT for platform events published by application code (use apex/platform-events-apex). NOT for external CDC subscribers via CometD or Pub/Sub API (use integration/change-data-capture-integration).

Use when writing, reviewing, or debugging test classes for Salesforce CPQ (Steelbrick) functionality — including quote creation, price rules, contracting, ordering, and CPQ API integration. Trigger keywords: CPQ test class, SBQQ test, quote calculation test, price rule test, CPQ Apex test, ServiceRouter test, QuoteCalculator test. NOT for standard Apex testing patterns unrelated to CPQ, nor for UI/Selenium test authoring outside the CPQ context.

Use when building or reviewing code that constructs SOQL/SOSL at runtime, inspects schema metadata via Schema.describe methods, accesses fields dynamically on sObjects, or performs runtime type inspection. Triggers: 'Database.query', 'Schema.getGlobalDescribe', 'Schema.describeSObjects', 'dynamic field access', 'SObjectType', 'DescribeFieldResult'. NOT for static SOQL queries or query performance tuning — use soql-fundamentals or soql-query-optimization.

Use when writing, reviewing, or debugging Apex exception handling, DmlException behavior, custom exception hierarchies, or user-safe error messages. Triggers: 'DmlException', 'swallowed exception', 'AuraHandledException', 'trigger rollback', 'try catch'. NOT for choosing async execution models or general governor-limit tuning.

Use this skill when designing or implementing FSC-specific integration: core banking data sync, custodian feeds (Schwab/Fidelity), market data pipelines, or payment processing wired to FSC Financial objects. Triggers: syncing FinancialAccount or FinancialHolding records from a core banking system, integrating a custodian data feed into FSC Wealth Management, market data prices updating CurrentValue on FinancialHolding, payment transactions flowing into FSC from an external ledger. NOT for generic Salesforce integration, non-FSC object sync, or MuleSoft platform configuration unrelated to FSC financial objects.

Use when writing, reviewing, or troubleshooting Apex that risks hitting Salesforce governor limits. Triggers: 'too many SOQL queries', 'too many DML statements', 'CPU time limit', 'bulkification', 'Queueable vs Batch'. NOT for trigger architecture decisions unless the core problem is limit consumption.

Use when designing or reviewing Apex actions exposed to Flow or similar orchestration layers via `@InvocableMethod`, especially around wrapper DTOs, bulk-safe list contracts, and error behavior. Triggers: 'InvocableMethod', 'InvocableVariable', 'Flow Apex action', 'bulk-safe invocable', 'Apex action input/output'. NOT for general Flow design without a custom Apex action boundary.

Use when designing or reviewing multi-step Apex orchestration that spans multiple transactions, checkpoints state across async boundaries, or recovers from partial failure in a long-running workflow. Trigger keywords: 'multi-step Queueable chain', 'cross-transaction state', 'platform event state machine', 'Finalizer retry', 'orchestrate across transactions', 'workflow progress tracking'. NOT for single-job Queueable design (use apex-queueable-patterns), NOT for Flow Orchestration (use flow/orchestration-flows), NOT for bulk record processing where Batch Apex is the primary tool (use batch-apex-patterns).

Use when building or reviewing Apex triggers on the Opportunity object — stage-change detection, amount rollups to Account, OpportunityTeamMember sync, or OpportunitySplit calculations. Trigger keywords: stage change automation, opportunity rollup, team member sync, split percentage, Trigger.oldMap stage comparison. NOT for generic trigger framework structure (use apex/trigger-framework) and NOT for Flow-based opportunity automation (use admin/flow-for-admins).

Use when designing or reviewing Salesforce Platform Cache usage in Apex, including org cache versus session cache, cache-aside patterns, invalidation, and safe key design. Triggers: 'Platform Cache', 'Cache.Org', 'Cache.Session', 'cache-aside', 'cache invalidation'. NOT for durable configuration storage or for caching user-specific sensitive data.

Use when writing, reviewing, or troubleshooting Apex queries that may expose SOQL injection or CRUD/FLS issues. Triggers: 'Database.query', 'WITH USER_MODE', 'WITH SECURITY_ENFORCED', 'stripInaccessible', 'security review finding'. NOT for record-sharing design unless the main issue is Apex query security.

Use this skill when writing, debugging, or reviewing Server-Side JavaScript (SSJS) in Salesforce Marketing Cloud — Script Activities, Cloud Pages, and Landing Pages. Covers WSProxy for SOAP API access, Script.Util.HttpRequest for outbound REST calls, error handling patterns, execution limits, and SSJS/AMPscript interoperability. NOT for AMPscript-only personalization logic inside Email Studio sends, not for standard Apex (Salesforce Platform), and not for client-side JavaScript in Experience Cloud or LWC.

Use when writing, reviewing, or debugging Apex test classes, test data factories, async test behavior, negative-path assertions, or callout mocking. Triggers: 'SeeAllData', 'Test.startTest', 'HttpCalloutMock', 'test data factory', 'missing assertions'. NOT for LWC Jest tests or code-coverage-only conversations detached from test design.

Use when writing or reviewing Apex, SOQL, or Flow logic that involves Date or Datetime fields: scheduling jobs, comparing dates across timezones, displaying Datetime values to users, filtering records by date literals, or persisting Datetime from external systems. NOT for calendar UI component layout, time-zone configuration in org settings, or Salesforce Connect external object date mapping.

Use when writing, reviewing, or designing Apex triggers. Triggers: 'trigger', 'trigger handler', 'trigger framework', 'recursion', 'before insert', 'after update', 'one trigger per object'. NOT for Flow-based automation — use admin/flow-for-admins for declarative automation decisions.

Use this skill to design and review the architecture for integrating external AI agents (Claude, ChatGPT, LangChain, custom LLM pipelines) with a Salesforce org — covering integration pattern selection (MCP vs. REST vs. Platform Events), auth model design, data exposure scope, org governor limit planning, and Well-Architected tradeoffs. Trigger keywords: AI agent Salesforce integration, agent-to-org architecture, LLM Salesforce access pattern, external AI org connectivity, agentic system org design. NOT for native Agentforce Agent creation inside Salesforce (use agent-topic-design), NOT for implementation-level Apex tool code (use mcp-tool-definition-apex), NOT for ETL or data pipeline design (use sf-to-llm-data-pipelines).

UX co-pilot — conversational UX designer with live preview. Discovery, audit, preview, export.

Use for hands-on Pulumi CLI work: running deployments, fixing broken stacks, and managing infrastructure state. Handles: recovering from stuck or interrupted `pulumi up` with pending operations, cleaning orphaned resources from state after out-of-band cloud deletions, protecting critical resources from accidental `pulumi destroy`, moving resources between stacks without recreating them, targeting specific resources during deployment, migrating between backends (local file to Pulumi Cloud, S3), stack lifecycle management, state export/import/repair, CI/CD pipeline setup, and importing existing cloud resources. Use this skill — not the language-specific Pulumi skills — whenever the user's question is about operating, troubleshooting, or recovering Pulumi infrastructure rather than writing program code.

go-code-review

go-code-audit

Step-by-step workflow to fix npm/pnpm/yarn vulnerabilities and review Dependabot PRs with semver and CI safety.

Generate comprehensive PR titles and descriptions with summary, test plan, risks, and linked issues. Use when the user asks to write or improve a pull request description.

vercel --version

'Merge overlapping review specialist agents into a single general-review-specialist.

Create pull requests with structured description

Create or update a GitHub pull request following Medox conventions (author spicode-bot, reviewer spideystreet). Invoke manually with /pr — do NOT trigger automatically.

Review recent changes and update CLAUDE.md to keep it relevant for future sessions. Use after significant changes or periodically to maintain context.

Generate an AI-synthesised summary of your GitHub activity (issues, PRs, commits) for a specified organisation and time period. Creates a natural language narrative with inline URL references. Use when user wants to summarise their work, review their contributions, or generate a work log.

Generate 5 distinct, production-grade UI variations for a requested new or existing section, component, page, flow, or shell, then help the user preview and apply one. Use when the user asks to add or redesign UI like a hero, pricing, navbar, auth flow, dashboard shell, ecommerce surface, or data component.

Use to update work item state on the project board at lifecycle phase transitions. Automatically invoked by /ai-brainstorm (refinement, ready), /ai-dispatch (in_progress), and /ai-pr (in_review). Also invoke manually for 'move this issue to in-review', 'update the board', 'mark as in progress', 'sync the work item state'. Fail-open: never blocks the calling workflow.

An experimental prompt engineering project. Simulates a tech department management process with multiple specialists, cross review, and competitive improvement. Default 3 person team, 11 person involution mode optional. Emergency exit always available. A fun exploration of LLM capability boundaries.

license-compliance

Survey recent work on this branch. Analyze commits, diffs, and code quality. Create beads for any deficiencies found.

Use when working with GitHub PRs, issues, workflows, or CI/CD - automates GitHub operations from terminal