kugamon

Run a comprehensive Salesforce org audit producing 18 documents from a single scan. Inventories and scores Apex classes, triggers, Flows, Process Builders, Workflow Rules, LWC components, custom objects/fields, validation rules, formula fields, approval processes, escalation/assignment/auto-response rules, Profiles, and Permission Sets. Detects unused custom fields and objects (no data and/or no references in code, flows, layouts, or reports). Also covers Reports & Dashboards, integrations, test coverage, licensing, team evaluation, change history, data quality, and architectural analysis. Scans all formula and criteria logic for hardcoded Record IDs, Campaign names, Profile names, URLs, and other fragile values. Generates Word, Excel, HTML reports plus 12 standalone analysis documents including a customer report and strategic engagement plan. Usage: /sf-audit [full|apex|flow|lwc|metadata|permissions|reports|integrations|coverage|licensing|team|change-history|data-quality] ...

Audits Salesforce orgs and codebases for security vulnerabilities — CRUD/FLS enforcement, SOQL injection, sharing violations, hardcoded secrets, unsafe Lightning patterns, and PII exposure — with a 100-point scored report and an AppExchange security review readiness checklist. Use when the user asks for a security audit, security review, vulnerability scan, AppExchange submission prep, CRUD/FLS check, or asks "is this code secure". Usage: /sf-security [audit|review|fix] [org|class|component] {name} ...

Captures and analyzes Salesforce debug logs through the Tooling API — trace flag setup, log retrieval, parsing for exceptions, SOQL-in-loop detection, CPU/heap/limit analysis, and row-lock diagnosis — all MCP-first against the live org. Use when the user hits an error in Salesforce, mentions debug logs, governor limits, "too many SOQL queries", CPU timeouts, UNABLE_TO_LOCK_ROW, flow errors, or asks why something failed in the org. Usage: /sf-debug [trace|logs|analyze|limits] {user|class|logId} ...

Generates, reviews, and runs Salesforce Apex test classes with a 120-point test-quality scoring rubric using a Salesforce MCP server. Use when writing test classes, improving code coverage, reviewing existing tests for assertion quality and bulk safety, running test suites, or diagnosing coverage gaps before a deployment or AppExchange submission. Usage: /sf-test [generate|review|run] [class|all] {name} ...

Generates and reviews Salesforce Apex code with best practices and 150-point scoring using a Salesforce MCP server. Use when writing Apex classes, triggers, test classes, batch jobs, or reviewing existing Apex code for bulkification, security, and SOLID principles. Usage: /sf-apex [create|update|validate] [class|trigger|test-class] {name} ...

Salesforce data and SOQL expert. Execute SOQL queries (natural language or raw SOQL), build optimized queries with selectivity analysis, insert/update/upsert/delete records, validate data operations, describe objects, and manage test data via Salesforce MCP server. Usage: /sf-data [query|build-query|insert|update|upsert|delete|validate|describe] {target} ...

Creates Salesforce architecture diagrams using Mermaid with ASCII fallback. Use when visualizing OAuth flows, data models (ERDs), integration sequences, system landscapes, role hierarchies, or Agentforce agent architectures. Usage: /sf-diagram [oauth|erd|integration|landscape|hierarchy|agentforce] ...

Creates and validates Salesforce flows with 110-point scoring and Winter '26 best practices using Salesforce MCP server. Use when building record-triggered flows, screen flows, autolaunched flows, scheduled flows, or reviewing existing flow performance. Usage: /sf-flow [create|update|validate] {FlowName} ...

Audit user-facing messages in a Salesforce managed package for grammatical issues and generate a prioritized remediation report. Extracts messages from Apex classes, Apex triggers, LWC, Aura Components, Visualforce Pages, and Visualforce Components — same extraction as sf-message-catalog — then runs each through grammar analysis to flag article errors, subject-verb agreement, plural/singular mismatches, spelling, capitalization, punctuation, and cross-component wording inconsistencies. Produces a .docx report with current message, issue, suggested fix, component reference, and priority. Use when the user mentions 'grammar audit', 'message quality', 'fix error messages', 'grammatical issues', 'proofread messages', 'message consistency', 'clean up error messages', 'improve error messages', or asks to review messages for grammar, spelling, or consistency. Also triggers on 'typos in our messages' or 'message style review'. Internal-facing; can be scheduled monthly.

Generate a comprehensive catalog of every user-facing error, warning, info, and confirmation message across a Salesforce managed package. Extracts messages from Apex classes, Apex triggers, Lightning Web Components (LWC), Aura Lightning Components, Visualforce Pages, and Visualforce Components. Produces a professional .docx document organized by component type with an executive summary table. Use this skill when the user mentions 'error catalog', 'message catalog', 'list all errors', 'list all warnings', 'user-facing messages', 'error messages document', 'warning messages report', 'managed package messages', 'Q2C messages', or asks to catalog, inventory, or document the error and warning messages in any Salesforce app. Also triggers on 'what messages does the app show', 'customer-facing errors', or 'generate message documentation'. This skill can be scheduled to run monthly to keep the catalog current across releases.

Manage the full Kugamon Quote-to-Cash lifecycle in Salesforce — opportunities, quotes, orders, invoices, payments, shipments, and assets, which uses the kugo2p namespace (Kugamon Quote to Cash). And optionally Managed the full Kugamon Subscription Billing lifecycle in Salesforce - opportunities, quotes, orders, invoices, payments, shipments, contracts, subscriptions, and assets, which requires the kuga_sub namespace (Kugamon Subscription Management). Detects which packages are installed and adapts accordingly. Use when users request operations on any Kugamon object.

Extract transcripts from YouTube videos. Use when the user asks for a transcript, subtitles, or captions of a YouTube video and provides a YouTube URL (youtube.com/watch?v=, youtu.be/, or similar). Supports output with or without timestamps.

Use the Perplexity MCP for any task that needs fresh / external / web-authoritative information. Triggers on requests like "research X", "look up Y", "what's happening with Z", "fact-check this", "find recent news about...", "competitive analysis of...", "what are customers saying about...", "is this still true?", "ask Perplexity...", "deep research on...", "do a teardown of [company]", or any question that benefits from cited live-web answers rather than knowledge from training. Choose the right Perplexity tool — quick lookup → perplexity_search; synthesized answer → perplexity_ask; multi-step analysis → perplexity_reason (Pro); long-form report → perplexity_research (Pro); ASI / heavy computation → perplexity_compute (Pro/Max). When Pro tools are unavailable on a Free account, chain multiple perplexity_search calls and synthesize manually — the source-finding works at every tier.