vibekit

Consolidate and synthesize research outputs from multiple AI models or sources into a unified, pattern-aware, provenance-enriched report with quality metrics. Use when the user has research outputs to consolidate, wants to synthesize multiple reports, asks to "consolidate", "synthesize", or "merge" research findings, or needs to reconcile conflicting information from different sources. Works with outputs from Claude, Gemini, GPT-5.2, or any combination of AI/human sources. Supports manifest-driven (from create-research-brief) and standalone operation.

Designs multi-model AI research strategies across 9 research patterns for Claude Opus 4.6, Gemini 3.1 Pro Deep Research, and optionally GPT-5.2 Deep Research (with site restrictions and mid-session intervention) or GPT-5.2 Chat. Generates copy-pasteable prompts optimized per model with pattern-aware role assignment, a consolidation manifest for downstream synthesis, and a merge prompt for consolidating outputs. Accepts Research Request Specification from research-interviewer upstream. Triggers on "create research brief", "research plan", "multi-model research", "research prompts for Claude/Gemini/OpenAI", "design research strategy", "consolidate research outputs", "synthesize research results", "research best practices for...", "best practices for [technology]", "create a best practices guide for...", "research [technology] patterns", "document [technology] best practices", "compare X vs Y", "landscape of [domain]", "compliance requirements for [topic]", "ROI analysis for [decision]". Modes: DUAL (Claude+Gemi

Generate reader-centered documentation aligned with the Diátaxis framework and optimized for cognitive load management. Creates tutorials, how-to guides, explanations, and reference documentation with multi-layer quality validation. PROACTIVELY activate for: (1) Write tutorials for learning new skills, (2) Create how-to guides for specific tasks, (3) Write explanations for understanding concepts, (4) Generate reference documentation for APIs or features, (5) Document systems, features, or processes, (6) Create user guides or onboarding documentation. Triggers: "write documentation for", "create docs", "document this", "write a tutorial for", "how-to guide for", "explain [concept]", "reference documentation for", "API docs for", "user guide for", "document the", "create documentation"

Systematic completeness audit of Architecture Documentation using 188-item viewpoint-based checklist, severity-classified gap detection, technical debt indicators, and architecture anti-pattern scanning. Supports TOGAF, C4, arc42, and IEEE 42010 frameworks. PROACTIVELY activate for: (1) Architecture review gates, (2) ADR validation before implementation, (3) C4 diagram completeness check, (4) Technical debt assessment, (5) Pre-implementation validation, (6) Governance compliance audit, (7) Design doc handoff review. Triggers: "audit architecture", "review ADR", "check architecture doc", "validate design doc", "architecture review", "audit C4 diagrams", "check system context", "technical debt assessment", "architecture health check", "governance review", "architecture completeness"

Systematically surface, classify, and stress-test assumptions in decisions, strategies, and plans. Transforms hidden assumptions into visible, testable propositions with load-bearing analysis and counterfactual validation. PROACTIVELY activate for: (1) Pre-commitment decision reviews, (2) Strategy validation before execution, (3) Investment due diligence, (4) Architecture decision records, (5) Product direction pivots, (6) Risk assessments requiring assumption audit. Triggers: "validate assumptions", "test assumptions", "assumption check", "stress test this decision", "what are we assuming", "pre-mortem", "what could go wrong", "challenge this plan", "devil's advocate"

Multi-expert comparison and ranking of options with structured deliberation. PROACTIVELY activate for: (1) decision making between alternatives, (2) trade-off analysis, (3) option evaluation, (4) vendor/tool comparison, (5) strategy selection. Triggers: "compare options", "evaluate alternatives", "rank options", "which should I choose", "compare these", "trade-off analysis", "decision between"

Systematic completeness audit of Product Requirements Documents using 100+ item MECE checklist, severity-classified gap detection, and anti-pattern scanning. PROACTIVELY activate for: (1) PRD review before development handoff, (2) Requirements completeness assessment, (3) Identifying ambiguous requirements, (4) PRD template validation, (5) Finding missing stakeholder needs, (6) Detecting inconsistent acceptance criteria. Triggers: "audit PRD", "review requirements", "check PRD completeness", "validate requirements document", "PRD review", "requirements audit", "find gaps in PRD", "requirements completeness check", "PRD quality check"

Iterative adversarial stress-testing through Red/Blue team dynamics. Red Team generates substantive, steel-manned attacks against propositions; Blue Team responds with defenses, mitigations, and hardening. Cycles continue until convergence criteria are met, producing a battle-tested proposition. PROACTIVELY activate for: (1) High-stakes decisions requiring stress-testing, (2) Strategy validation before major commitment, (3) Architecture decision hardening, (4) Proposal defense preparation, (5) Security posture review, (6) Investment due diligence with adversarial lens. Triggers: "red team this", "blue team", "stress test", "attack this plan", "find weaknesses", "adversarial review", "devil's advocate", "what could go wrong", "poke holes in this", "challenge this decision", "war game this"

Prompt engineering architect that transforms raw prompts into optimized, production-ready prompts. PROACTIVELY activate for: (1) Improve/refine prompts, (2) Optimize prompts for specific models, (3) Create prompt chains/sequences, (4) Analyze prompt quality, (5) Transform prompts for different contexts. Triggers: "improve this prompt", "refine prompt", "optimize prompt", "make this prompt better", "prompt engineering", "fix my prompt"

Systematic knowledge elicitation through structured interviewing with epistemic confidence tracking, MECE coverage verification, and bias-protected questioning. PROACTIVELY activate for: (1) Gather research requirements, (2) Elicit problem statements, (3) Extract domain knowledge, (4) Clarify research goals, (5) Generate requirements through discovery. Triggers: "interview me", "elicit knowledge", "extract information", "research interview", "gather requirements", "conduct interview", "knowledge extraction"

Foundational knowledge for creating ADK (Agent Development Kit) agents including environment setup, project structure, and basic agent scaffolding. PROACTIVELY activate for: (1) new ADK agent creation, (2) ADK project setup and environment configuration, (3) AdkApp initialization and understanding core ADK architecture. Triggers: "create adk agent", "new agent", "setup adk"

Multi-agent orchestration patterns including coordinator agents, LlmAgent vs WorkflowAgent selection, agent-as-tool pattern, and inter-agent communication via session state. PROACTIVELY activate for: (1) multi-agent systems and agent coordination, (2) sub-agent delegation and agent-as-tool implementation, (3) workflow orchestration with LlmAgent and WorkflowAgent. Triggers: "multi-agent", "orchestration", "agent team"

Best practices for engineering high-quality system prompts for AI agents with emphasis on XML structure, clarity, few-shot examples, and token efficiency. PROACTIVELY activate for: (1) system prompt creation and optimization, (2) instruction clarity improvement and few-shot example design, (3) XML tag structure implementation and prompt templates. Triggers: "system prompt", "optimize prompt", "prompt engineering"

Vertex AI RAG Engine integration patterns for grounding agent responses in private data sources including corpus management, retrieval tool creation, and citation extraction. PROACTIVELY activate for: (1) RAG pipeline integration and Vertex AI RAG Engine setup, (2) corpus creation and document ingestion, (3) retrieval tool configuration and grounding metadata parsing. Triggers: "add rag", "rag pipeline", "vertex ai search"

Vertex AI SDK patterns for configuring Gemini models including generation parameters, safety settings, streaming responses, and function calling. PROACTIVELY activate for: (1) model configuration and temperature settings, (2) safety controls and streaming implementation, (3) Vertex AI endpoint selection and function calling. Triggers: "gemini config", "model parameters", "streaming response"

[STUB - Not implemented] Lightweight architectural decision tracking in Markdown (ADR) format. PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

[STUB - Not implemented] C4 model diagrams for layered architecture visualization (Context, Container, Component, Code). PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

[STUB - Not implemented] Entity-relationship diagrams and data modeling for database schema design. PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

[STUB - Not implemented] Mermaid.js diagrams-as-code for version-controlled, reviewable visual documentation. PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

[STUB - Not implemented] OpenAPI 3.1.0 design-first API specifications as single source of truth. PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

[STUB - Not implemented] Sequence diagram patterns for documenting system interactions and API flows. PROACTIVELY activate for: [TODO: Define on implementation]. Triggers: [TODO: Define on implementation]

Architecture Decision Records (ADRs) format, templates, and best practices for documenting technical decisions. PROACTIVELY activate for: (1) documenting significant technical decisions, (2) recording design trade-offs, (3) creating technology selection rationales. Triggers: "ADR", "decision record", "architecture"

Server-side Firebase Admin SDK patterns for Next.js 14+ with secure initialization and token verification. PROACTIVELY activate for: (1) setting up Admin SDK with server-only protection, (2) implementing custom claims for RBAC, (3) verifying tokens in middleware. Triggers: "admin sdk", "server-only", "custom claims"

Firebase Authentication implementation patterns including sign-in flows, OAuth, and session management. PROACTIVELY activate for: (1) implementing sign-in/sign-up flows, (2) setting up OAuth providers like Google, (3) creating protected routes with middleware. Triggers: "auth", "sign-in", "oauth"

Next.js 14+ App Router integration with Firebase including server/client SDK separation and data fetching patterns. PROACTIVELY activate for: (1) setting up dual SDK architecture for client/server, (2) fetching data in Server Components with Admin SDK, (3) implementing real-time listeners in Client Components. Triggers: "nextjs firebase", "app router", "server client sdk"

Firestore data modeling best practices including subcollections, document structure, and atomic operations. PROACTIVELY activate for: (1) designing Firestore collection structures, (2) choosing between subcollections vs root collections, (3) implementing transactions vs batched writes. Triggers: "subcollection", "data model", "firestore"

Firestore Security Rules patterns for user-scoped access, RBAC, and field validation. PROACTIVELY activate for: (1) implementing user-scoped data access rules, (2) setting up role-based access with custom claims, (3) validating fields and enforcing immutability. Triggers: "security rules", "rbac", "firestore rules"

Framer Motion animations including motion components, gestures, entrance/exit, and scroll triggers. PROACTIVELY activate for: (1) adding hover and tap animations, (2) implementing entrance/exit transitions with AnimatePresence, (3) creating scroll-triggered animations. Triggers: "animate", "framer motion", "whileHover"

Data fetching patterns for Next.js 14+ App Router including async Server Components and Server Actions. PROACTIVELY activate for: (1) fetching data in Server Components, (2) implementing Server Actions for mutations, (3) configuring caching and revalidation. Triggers: "fetch data", "server action", "revalidate"

Special file conventions in Next.js 14+ App Router for pages, layouts, and API routes. PROACTIVELY activate for: (1) creating pages and layouts, (2) implementing loading and error states, (3) building API route handlers. Triggers: "new page", "layout", "api route"

Opinionated Next.js 14+ project structure and configuration for scaffolding new projects. PROACTIVELY activate for: (1) setting up new Next.js projects, (2) configuring tsconfig.json with strict mode, (3) establishing folder structure patterns. Triggers: "new project", "scaffold", "project structure"

React Server Components vs Client Components decision framework with composition patterns. PROACTIVELY activate for: (1) deciding when to use 'use client' directive, (2) implementing island architecture with small client leaves, (3) passing server data to client components. Triggers: "server component", "client component", "use client"

React Hook Form useFieldArray for dynamic list management with Zod validation. PROACTIVELY activate for: (1) creating dynamic form arrays with add/remove, (2) validating arrays with Zod min/max, (3) using field.id as key. Triggers: "field array", "useFieldArray", "dynamic form"

React Hook Form v7 with zodResolver integration for type-safe validated forms. PROACTIVELY activate for: (1) creating forms with useForm and zodResolver, (2) implementing field registration with spread syntax, (3) handling form state and errors. Triggers: "react hook form", "useForm", "zodResolver"

React Server Components architecture patterns for Next.js 14+ with server-first development. PROACTIVELY activate for: (1) deciding Server vs Client Component boundaries, (2) composing components with 'use client', (3) managing interactivity patterns. Triggers: "server component", "client component", "'use client'"

shadcn/ui CLI usage, component installation, and composition patterns with file ownership philosophy. PROACTIVELY activate for: (1) adding shadcn/ui components via CLI, (2) composing multi-part components, (3) customizing component variants. Triggers: "shadcn", "add component", "components.json"

Tailwind CSS utility-first styling with mobile-first responsive design and CSS variables. PROACTIVELY activate for: (1) implementing mobile-first responsive layouts, (2) using semantic CSS variables for theming, (3) creating component variants with cva. Triggers: "tailwind", "responsive", "dark mode"

TypeScript interface conventions for component props with JSDoc comments and utility types. PROACTIVELY activate for: (1) defining component prop interfaces, (2) using generics for reusable components, (3) integrating cva VariantProps. Triggers: "props", "interface", "VariantProps"

TypeScript patterns for type-safe API contracts with strict typing and Zod integration. PROACTIVELY activate for: (1) designing type-safe API interfaces, (2) creating Zod schemas for validation, (3) implementing generic response wrappers. Triggers: "api design", "type safety", "zod validation"

Web accessibility patterns including semantic HTML, ARIA attributes, and keyboard navigation. PROACTIVELY activate for: (1) using semantic HTML elements correctly, (2) adding ARIA labels for icons and dynamic content, (3) implementing keyboard navigation. Triggers: "accessibility", "a11y", "aria"