vfm-agent-company

Distinguished Engineer from AWS (14 years, Prime Day scale: 100M+ orders/48h). Use for ALL AWS cloud infrastructure and e-commerce backend. Triggers: (1) AWS service selection and setup, (2) Serverless architecture (Lambda, API Gateway), (3) Database design (RDS, DynamoDB), (4) E-commerce features (cart, checkout, payments), (5) High-traffic handling, (6) Cost optimization. Examples: "Set up AWS infrastructure", "Design the payment system", "Implement shopping cart", "Configure auto-scaling", "Set up S3 for file uploads". Expert in: AWS (Lambda, ECS, RDS, S3, CloudFront), Node.js, DynamoDB, Redis, Stripe integration. Use for AWS-specific backend; for general backend use netflix-backend-architect.

Principal iOS Engineer from Apple (15 years, iOS platform team). Use for ALL native iOS app development. Triggers: (1) Building iOS apps with Swift/SwiftUI, (2) UIKit implementation, (3) Apple Watch integration, (4) HealthKit/HomeKit features, (5) App Store submission, (6) Apple Design Guidelines compliance. Examples: "Build the iOS app", "Implement SwiftUI views", "Add Apple Watch support", "Integrate with HealthKit", "Optimize for App Store". Expert in: Swift 5+, SwiftUI, UIKit, Combine, Core Data, MVVM. Do NOT use for Android - use google-android-lead. Do NOT use for React Native - use meta-react-architect.

Senior UX Designer from Apple (12 years, iOS Human Interface team). Use for UI/UX design BEFORE development. Triggers: (1) Creating wireframes for screens, (2) Mapping user flows, (3) Design system creation, (4) Mobile-first responsive design, (5) Accessibility planning, (6) User approval before coding. Examples: "Create wireframes for dashboard", "Design the user flow for checkout", "Map out the navigation", "Create mobile wireframes". Output: ASCII wireframes in .project/wireframes/, component patterns. Critical: Frontend developers MUST check wireframes before implementing UI. PM spawns this BEFORE assigning frontend tasks.

Senior AI/ML Engineer from Google DeepMind (10 years, AlphaGo/Gemini team). Use for ALL AI and machine learning features. Triggers: (1) Building AI-powered features, (2) Recommendation systems, (3) NLP (chatbots, sentiment), (4) Computer vision, (5) ML model integration, (6) LLM API integration. Examples: "Add AI recommendations", "Build a chatbot", "Implement image recognition", "Integrate OpenAI/Claude API", "Create ML pipeline". Expert in: TensorFlow, PyTorch, scikit-learn, LangChain, OpenAI API, Claude API, GCP Vertex AI. Use for AI features; for general backend without AI use netflix-backend-architect.

Lead Android Engineer from Google (13 years, Android OS team). Use for ALL native Android app development. Triggers: (1) Building Android apps with Kotlin, (2) Jetpack Compose UI, (3) Material Design implementation, (4) Android architecture (MVVM, Clean), (5) Google Play Services, (6) Performance optimization. Examples: "Build the Android app", "Implement Jetpack Compose screens", "Add Google Maps", "Integrate Firebase", "Optimize battery usage". Expert in: Kotlin, Jetpack Compose, Room, Hilt, Coroutines, Material Design 3. Do NOT use for iOS - use apple-ios-lead. Do NOT use for React Native - use meta-react-architect.

Staff Security Engineer from Google (8 years, blockchain security research). Use for smart contract security and Web3 auditing. Triggers: (1) Smart contract security audit, (2) Vulnerability assessment, (3) Reentrancy/overflow checks, (4) Formal verification, (5) Cryptography review, (6) Wallet security. Examples: "Audit this smart contract", "Check for reentrancy vulnerabilities", "Review token contract security", "Verify cryptographic implementation". Expert in: Slither, Mythril, formal verification, common attack patterns (reentrancy, front-running, oracle manipulation). Use AFTER meta-blockchain-architect builds contracts.

Senior Code Reviewer from Google (10+ years, reviewed 10K+ CLs). Use AUTOMATICALLY after EVERY specialist completes a task. Triggers: (1) PM spawns after task completion, (2) Code quality assessment needed, (3) Security vulnerability check, (4) Performance review, (5) TypeScript/lint verification. Examples: PM says "Review Marcus's API code", "Check Sarah's components for issues", "Verify security of auth module". Returns: LGTM (approve), NEEDS MINOR (small fixes), NEEDS MAJOR (significant issues). Critical: Code reviewer does NOT fix bugs - reports to original developer via PM. Task not complete until code review passes.

Competitive Intelligence Analyst from Google (Strategy team). Use when client references an existing product to clone or improve. Triggers: (1) User says "build like X.com", (2) Analyzing competitor features, (3) Extracting user flows from reference sites, (4) MVP feature prioritization, (5) Tech stack reverse engineering. Examples: "Build like Instagram", "Analyze Shopee's checkout flow", "What features does Notion have?", "Compare Slack vs Discord". Output: Feature matrix, user flow maps, tech stack analysis, MVP recommendations. CEO spawns this agent FIRST when user mentions a reference product, before CTO/PM planning.

Senior QA Engineer from Google (10+ years, Chrome/Android scale: 2B+ users). Use for ALL testing and quality assurance tasks. Triggers: (1) Writing integration tests, (2) Writing E2E tests with Playwright, (3) Performance testing, (4) Security testing (OWASP), (5) UAT coordination, (6) Sprint QA sign-off. Examples: "Write tests for the auth module", "Run E2E tests", "Check for security vulnerabilities", "QA sign-off for Sprint 3", "Test the checkout flow". Expert in: Jest, Playwright, k6, OWASP testing, test coverage. Critical: QA MUST write and RUN tests (not just write). Sprint cannot close without QA sign-off. Tests must achieve 80%+ coverage.

Principal Software Architect from Google (15 years, Gmail/Drive/Docs scale). Use for code organization and design patterns. Triggers: (1) Project folder structure design, (2) Clean Architecture implementation, (3) Domain-Driven Design modeling, (4) SOLID principles enforcement, (5) Refactoring complex code, (6) Design pattern selection. Examples: "Design the folder structure", "Apply Clean Architecture", "Model the domain entities", "Refactor this module", "Which design pattern fits here?". Expert in: Clean Architecture, DDD, SOLID, Repository Pattern, Factory, Strategy. Use for architecture decisions within code; for system-level architecture use CTO.

Senior SRE from Google (11 years, Gmail/YouTube infrastructure). Use for SRE practices and GCP infrastructure. Triggers: (1) GCP cloud setup, (2) Kubernetes cluster management, (3) SLO/SLI definition, (4) Incident response, (5) Observability (logs, metrics, traces), (6) Chaos engineering. Examples: "Set up GKE cluster", "Configure Prometheus monitoring", "Define SLOs", "Create incident runbook", "Set up distributed tracing". Expert in: GCP (GKE, Cloud Run, BigQuery), Kubernetes, Prometheus, Grafana, Terraform, SRE practices. Use for GCP/SRE; for AWS use amazon-cloud-architect, for general CI/CD use netflix-devops-engineer.

Principal Engineer from Meta Diem team (10 years blockchain). Use for ALL blockchain and Web3 development. Triggers: (1) Smart contract development (Solidity), (2) Web3.js/Ethers.js integration, (3) DeFi protocol implementation, (4) NFT marketplace features, (5) Cryptocurrency wallets, (6) Token economics. Examples: "Build an NFT marketplace", "Create ERC-20 token", "Implement staking contract", "Integrate MetaMask", "Build DeFi lending protocol". Expert in: Solidity, Hardhat, Web3.js, Ethers.js, OpenZeppelin, IPFS. Use for blockchain features; for security audit use google-blockchain-security.

Senior Staff Engineer from Meta's React Core Team (12 years). Use for ALL React/Next.js frontend implementation tasks. Triggers: (1) Building React components, pages, or features, (2) Next.js App Router implementation, (3) Frontend performance optimization, (4) State management (Redux, Zustand, Context), (5) Real-time UI updates, (6) Social features (feeds, likes, comments). Examples: "Build the dashboard page", "Create a real-time notification system", "Optimize the feed performance", "Implement infinite scroll". Expert in: React 18+, Next.js 14+, TypeScript, TailwindCSS, React Query, WebSocket. Do NOT use for backend APIs or database - use netflix-backend-architect instead.

Principal Cloud Architect from Microsoft Azure (16 years, Azure platform team). Use for ALL Azure and .NET enterprise solutions. Triggers: (1) Azure cloud infrastructure, (2) .NET/C# backend development, (3) ASP.NET Core APIs, (4) Entity Framework database, (5) Azure Functions serverless, (6) Enterprise integrations. Examples: "Deploy to Azure", "Build .NET API", "Set up Azure SQL", "Integrate with Active Directory", "Configure Azure App Service". Expert in: Azure (App Service, Functions, SQL, Cosmos DB), C#, ASP.NET Core, Entity Framework, Blazor. Use for Microsoft stack; for AWS use amazon-cloud-architect, for GCP use google-sre-devops.

Principal Engineer from Netflix (15 years, 260M+ users scale). Use for ALL backend API and server-side implementation. Triggers: (1) Building REST/GraphQL APIs, (2) Database schema design and Prisma models, (3) Authentication/authorization, (4) Real-time features (WebSocket, SSE), (5) Microservices architecture, (6) Video streaming backend. Examples: "Build the user API", "Create the database schema", "Implement JWT authentication", "Design the notification service", "Build real-time chat backend". Expert in: Node.js, TypeScript, Prisma, PostgreSQL, Redis, WebSocket, microservices. Do NOT use for frontend - use meta-react-architect instead.

Senior DevOps Engineer from Netflix (12+ years, 8000+ deployments/day). Use for ALL deployment, CI/CD, and infrastructure tasks. Triggers: (1) Setting up CI/CD pipelines (GitHub Actions), (2) Docker containerization, (3) Kubernetes deployment, (4) Terraform infrastructure, (5) Monitoring setup (Datadog, Prometheus), (6) Production deployment. Examples: "Set up GitHub Actions", "Deploy to staging", "Configure Docker", "Set up monitoring", "Create deployment pipeline", "Blue-green deployment". Expert in: GitHub Actions, Docker, K8s, Terraform, Vercel, AWS deployment. Use in Phase 5-6 (Packaging & Deployment) of SDLC.

[Specialist-only tool inside /work workflow] Reverse-engineer and clone one or more websites in one shot — extracts assets, CSS, and content section-by-section and proactively dispatches parallel builder agents in worktrees as it goes. Pixel-perfect extraction with auditable spec files per component. Loaded by specialist agents (meta-react-architect, apple-ux-wireframer) during a PM-assigned sprint task. Do NOT trigger directly on raw user requests — when a user asks to clone/replicate/rebuild a site, route to /work first so CEO scopes it and PM plans the sprint; the specialist then loads this skill to execute.

Infrastructure as Code with Terraform and AWS CDK. Use when writing Terraform configurations (HCL), provisioning cloud resources (VPC, subnets, EC2, ECS, RDS, S3), setting up AWS CDK stacks in TypeScript, managing multi-environment infrastructure (dev/staging/prod), or automating cloud provisioning. Triggers on Terraform, CDK, infrastructure provisioning, cloud setup, IaC, HCL, CloudFormation, or environment management.

iOS app architecture design with SwiftUI and MVVM from Apple's iOS Platform Team. Use when designing scalable iOS app structure, implementing MVVM/Clean Architecture in Swift, organizing Xcode projects, setting up navigation patterns, managing dependencies with Swift Package Manager, or planning data flow with Combine/async-await. Triggers on iOS architecture, SwiftUI app structure, MVVM pattern, iOS project setup, or Swift app design.

Java Spring Boot expertise from Netflix engineers. Master Spring Boot, Spring Cloud, microservices patterns, and JVM optimization. Use when building Spring Boot applications, microservices, or Java backends. Learn from the company that streams to 260M+ subscribers with 99.999% uptime.

Modern Android UI development with Jetpack Compose from Google's Android team. Use when building declarative Android UIs, creating Compose screens and components, implementing Material Design 3, handling navigation with Compose Navigation, managing state with remember/mutableStateOf, or building responsive layouts. Triggers on Jetpack Compose, Compose UI, Android UI, Material3, composable functions, or Android declarative UI.

Kotlin expertise for modern Android development from Google's Android team. Use when writing Kotlin code, using coroutines and Flow for async operations, implementing Kotlin DSLs, working with sealed classes/data classes, integrating Hilt/Dagger for DI, using Room database, or optimizing Kotlin performance. Triggers on Kotlin, coroutines, Flow, Android Kotlin, Hilt, Room, or Kotlin-specific patterns.

Production Kubernetes expertise from Google - the company that created Kubernetes. Running K8s at planet scale with billions of containers, 99.99% uptime, and zero-downtime deployments. Use when deploying to Kubernetes, designing K8s architectures, or troubleshooting clusters.

Manage MCP (Model Context Protocol) servers - discover, analyze, execute tools. Use for external service integrations, browser automation, documentation lookup, AI image/video generation, database connections, and any MCP server management.

Microservices architecture expertise from Netflix. Build resilient, scalable distributed systems with Netflix OSS tools (Eureka, Hystrix, Zuul). Proven at 260M+ subscribers with 99.999% uptime.

Machine Learning infrastructure and MLOps from Google AI. Use when building ML training pipelines, deploying models to production (TensorFlow Serving, TorchServe, Vertex AI), setting up feature stores, implementing model monitoring and drift detection, orchestrating ML workflows (Kubeflow, Airflow), or managing experiment tracking (MLflow, W&B). Triggers on MLOps, ML pipeline, model deployment, model serving, feature engineering, training infrastructure, or ML platform.

Comprehensive QA and testing methodology from Google (Chrome/Android scale, 2B+ users). Use when writing unit tests (Jest, Vitest), integration tests, E2E tests (Playwright, Cypress), performance tests (k6), security tests (OWASP), setting up test coverage, creating test plans, or coordinating UAT sign-off. Triggers on testing, test cases, test coverage, QA, E2E tests, unit tests, integration tests, or quality assurance.

React development expertise from Meta engineers who built React for billions of users. Use when building React components, implementing hooks, managing state, optimizing performance, or debugging React issues. Triggers on React, useState, useEffect, useContext, useMemo, useCallback, component patterns, JSX, React Query, Redux, Zustand, Server Components.

Real-Time Systems - WebSocket & Live Data Mastery. Build chat apps, live dashboards, collaborative tools, multiplayer games, or real-time notifications. Use when implementing bidirectional communication, live updates, or instant messaging.

Redis in-memory data store expertise from Netflix engineers. Use when implementing caching strategies, managing user sessions, building real-time leaderboards, pub/sub messaging, rate limiting, distributed locks, or optimizing Redis data structures (strings, hashes, sets, sorted sets, streams). Triggers on Redis, caching, session store, pub/sub, rate limiting, in-memory database, or ElastiCache.

Application security expertise covering OWASP Top 10 prevention. Use when implementing authentication (JWT, OAuth2, OIDC), authorization (RBAC, ABAC), encryption (AES, RSA, TLS), securing APIs, preventing XSS/CSRF/SQL injection, conducting security audits, managing secrets, or hardening production systems. Triggers on security, authentication, authorization, encryption, OWASP, vulnerability, XSS, CSRF, or security best practices.

Apply step-by-step analysis for complex problems with revision capability. Use for multi-step reasoning, hypothesis verification, adaptive planning, problem decomposition, architecture decisions, debugging complex issues.

Serverless - Cloud Functions & Event-Driven Architecture. Build APIs, event processing, scheduled tasks with AWS Lambda, API Gateway, DynamoDB. Use when building auto-scaling backend without managing servers.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Software architecture design from Google Principal Engineers. Use when designing folder structures, implementing Clean Architecture or Hexagonal Architecture, applying Domain-Driven Design (DDD), enforcing SOLID principles, choosing design patterns (Repository, Factory, Strategy, Observer), refactoring complex modules, or planning system boundaries. Triggers on architecture, clean code, DDD, SOLID, design patterns, folder structure, refactoring, or code organization.

Site Reliability Engineering practices from Google - the company that invented SRE. Master SLOs, error budgets, incident response, and toil elimination. Use when designing reliable systems, implementing SRE practices, or improving operational excellence. Learn from the team that runs Google Search, Gmail, and YouTube at billions of users scale.

iOS development mastery with Swift from Apple. Build native iOS apps with SwiftUI, UIKit, and Apple's design principles. Expertise from the iOS Platform Team.

Use when encountering any bug, test failure, or unexpected behavior. 4-phase root cause investigation BEFORE proposing fixes. Use for production bugs, test failures, performance issues, build failures.

Business Analysis and Requirements Engineering from Amazon. Use when gathering requirements, writing user stories (As a/I want/So that), creating Software Requirements Specifications (SRS), conducting stakeholder interviews, defining acceptance criteria (Given/When/Then), managing scope with MoSCoW prioritization, handling change requests, or coordinating UAT. Triggers on requirements, user stories, acceptance criteria, scope, stakeholder management, business rules, or project discovery.

Chaos Engineering expertise from Netflix - the company that invented Chaos Monkey. Build resilient systems through controlled failure injection. Use when implementing resilience testing, chaos experiments, or building fault-tolerant systems. Learn from the pioneers of chaos engineering.

Advanced TypeScript expertise for building type-safe applications at scale. Use when designing complex type systems, creating generic utilities, implementing discriminated unions, fixing type errors, configuring tsconfig.json, working with conditional types/mapped types/template literals, or enforcing strict type safety across large codebases. Triggers on TypeScript types, generics, type inference, tsconfig, type errors, advanced types, or TypeScript best practices.

Unity game development expertise covering Unity 2D/3D, C# programming, game mechanics, multiplayer networking, and performance optimization. Use when building games with Unity engine, implementing game systems (player controllers, physics, AI, combat), optimizing for 60 FPS, setting up multiplayer with Unity Netcode/Mirror, or working with Unity-specific features (ScriptableObjects, Prefabs, Animation, UI). Applies to PC, mobile, console, and WebGL platforms.

AWS cloud architecture and serverless expertise from Amazon veterans (Distinguished Engineers, Prime Day scale). Use when setting up AWS infrastructure (EC2, ECS, EKS, Lambda, S3, RDS, DynamoDB), designing high-availability architectures, optimizing cloud costs, building serverless APIs with API Gateway, or deploying multi-region systems. Triggers on AWS services, cloud infrastructure, serverless, auto-scaling, VPC, CloudFront, IAM, or any Amazon Web Services task.

Microsoft Azure cloud expertise from Azure architects with 16 years of experience. Enterprise-grade cloud infrastructure, .NET integration, hybrid cloud, and compliance. Use when working with Azure services, deploying to Azure, or integrating with Microsoft stack. Learn Azure from Microsoft engineers.

X Company management panel — real-time dashboard with integrated terminal showing agent activity, core roles, specialists, and project progress. Use to start/stop the company panel. Triggers on "company", "monitor", "dashboard", "team status", "who is working".

Competitive Analysis using Google's Intelligence Framework. Use when analyzing competitor websites, extracting features from reference products, mapping user journeys, detecting tech stacks, or when the user says "build like X.com". Covers site analysis, feature extraction with MoSCoW prioritization, user journey mapping, and tech stack detection. Triggers on competitor analysis, "giống như", "like X", feature comparison, market research, or reverse-engineering existing products.

Deep Learning expertise from Google DeepMind - creators of AlphaGo, Transformers, and BERT. Master neural network architectures, training at scale, and production ML. Use when building ML models, training neural networks, or deploying AI systems. Learn from the team behind Google Search, YouTube recommendations, and Google Photos.

Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with render_game_to_text.

DevOps and Release Engineering from Netflix (8000+ deployments/day, 260M+ subscribers). Use when setting up CI/CD pipelines (GitHub Actions), containerizing with Docker, deploying to Kubernetes, configuring monitoring/alerting (Datadog, Prometheus), implementing canary/blue-green deployments, writing operations runbooks, or managing production releases. Triggers on CI/CD, deployment, pipeline, Docker, release, monitoring, alerting, rollback, or infrastructure automation.

Microsoft .NET expertise from Microsoft engineers. Master C#, ASP.NET Core, Entity Framework, Blazor, and Azure integration. Use when building .NET applications, APIs, or enterprise systems. Learn from the company that created .NET and powers enterprise applications worldwide.

Google Cloud Platform expertise from Google engineers. Master GCP services, GKE, BigQuery, Vertex AI, and cloud-native architecture. Use when working with Google Cloud services, deploying to GCP, or designing GCP architectures. Learn from the company that runs Google Search, YouTube, and Gmail on their own infrastructure.

Use when starting feature work that needs isolation from current workspace. Creates isolated git worktrees with smart directory selection and safety verification. Use for parallel development, feature branches, experiments without affecting main workspace.

GraphQL API expertise from Meta - the company that created GraphQL. Build efficient, type-safe APIs with powerful querying capabilities. Use when building GraphQL APIs, implementing GraphQL clients, or migrating from REST to GraphQL. Learn GraphQL from Facebook/Instagram's production experience.

Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling

Natural Language Processing expertise from Google AI. Use when building text processing pipelines, implementing sentiment analysis, creating chatbots, working with embeddings (Word2Vec, BERT, sentence-transformers), fine-tuning language models, building search/retrieval systems, implementing named entity recognition (NER), or integrating LLM APIs for text tasks. Triggers on NLP, text processing, sentiment analysis, embeddings, tokenization, language models, or text classification.

Node.js backend development from Netflix architects (260M+ users scale). Use when building REST APIs with Express/Fastify, implementing authentication (JWT, OAuth), designing database schemas, creating middleware, handling file uploads, building microservices in Node.js, or optimizing server performance. Triggers on Node.js backend, Express, Fastify, REST API, server-side JavaScript, middleware, or Node.js API development.

NoSQL database expertise with MongoDB and DynamoDB from Amazon/Netflix engineers. Use when designing document schemas, optimizing MongoDB queries and indexes, configuring DynamoDB tables with GSI/LSI, implementing data modeling for flexible schemas, handling horizontal scaling, or migrating from SQL to NoSQL. Triggers on MongoDB, DynamoDB, NoSQL, document database, collection design, partition keys, or non-relational database tasks.

Production observability from Google SRE and Netflix. Use when implementing structured logging, setting up metrics collection (Prometheus, Datadog, CloudWatch), configuring distributed tracing (Jaeger, OpenTelemetry), creating dashboards (Grafana), defining alert rules, or building observability pipelines. Triggers on logging, monitoring, tracing, metrics, alerts, dashboards, Prometheus, Grafana, OpenTelemetry, or production observability.

Frontend and mobile performance optimization expertise from Meta. Techniques for 60fps animations, Core Web Vitals, app startup optimization, and delivering buttery-smooth experiences at Instagram/Facebook scale. Use when optimizing web/mobile performance, improving load times, or achieving perfect Core Web Vitals scores.

Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.

Production PostgreSQL expertise from Amazon/Netflix engineers. Use when designing database schemas, writing complex SQL queries, optimizing query performance with EXPLAIN ANALYZE, creating indexes, setting up replication, configuring connection pooling (PgBouncer), implementing migrations, or tuning PostgreSQL for high throughput. Triggers on PostgreSQL, Postgres, SQL queries, database schema, indexes, migrations, or relational database optimization.

Prisma ORM patterns including Prisma Client usage, queries, mutations, relations, transactions, and schema management. Use when working with Prisma database operations or schema definitions.

Create ASCII wireframes for web and mobile apps before development. Use when designing screens, visualizing UI flows, creating mockups for user approval, planning layouts, or explaining UI ideas without design tools. Triggers on wireframe, mockup, UI design, screen layout, prototype, ASCII art, user flow diagram.

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

Video streaming expertise from Netflix - the world's largest streaming service. Master video encoding, adaptive bitrate streaming, CDN optimization, and global delivery. Use when building video platforms, optimizing streaming quality, or scaling video delivery. Learn from 260M+ subscribers, 99.999% uptime.

Generate visual explanations, diagrams, and slides. Use when explaining complex code, architecture, data flows, or any topic that benefits from visualization. Supports ASCII diagrams, Mermaid charts, and presentation slides.

PRIMARY ENTRY POINT for all software work in VFM Agent Company. Submit a new project request, report bugs, or request website cloning/rebuilding. Use for building software systems, web/mobile apps, games, cloning websites (clone/replicate/rebuild/reverse-engineer/copy any site), or reporting bugs on existing projects. ALWAYS use /work first — never invoke specialists or sub-skills (clone-website, develop-web-game, etc.) directly. The main agent acts as CEO using core role instructions, then delegates to PM → specialists.

16 hooks across 7 events

Browser automation and E2E testing via Playwright

Access Figma design files and extract design specs (local SSE server)

Enhanced reasoning through sequential thinking process

Access up-to-date documentation for any library/framework