Production chat interfaces need authentication, persistence, context, and reliable backend integration. This skill provides focused React, Python, and Next.js implementation patterns.

Building ChatGPT Apps requires specific widget metadata, server structure, and testing steps. This skill provides templates and guidance for creating interactive Apps with MCP servers and embedded widgets.

Manual browser testing and data collection can be slow and hard to repeat. This skill gives Claude, Codex, and Claude Code a local Playwright MCP workflow for controlled browser actions.

New endpoints can become inconsistent with Catalyst-Relay conventions. This skill guides route creation, validation, registration, response formatting, client mapping, and documentation.

New core functions can become inconsistent with Catalyst-Relay architecture and result conventions. This skill guides file placement, imports, return types, helpers, and exports.

Static types alone cannot verify data arriving at application boundaries. This skill shows how Zod, tRPC, Prisma, and TypeScript connect compile-time types with runtime validation.

Manual tRPC setup creates repetitive router, schema, and registration work. This skill scaffolds consistent tRPC and Zod structures and checks basic project integration.

Unclear test priorities leave critical behavior exposed and waste automation effort. This skill creates risk-based plans, coverage targets, test cases, and CI quality gates.

Real-time APIs often fail under disconnects, slow consumers, and malformed stream handling. This skill provides focused SSE, WebSocket, ReadableStream, and LLM streaming patterns.

Server and client boundaries are difficult to design correctly in modern React applications. This skill provides Next.js patterns, templates, examples, and migration guidance.

Manual checks miss project-specific TypeScript and architecture violations. This skill runs focused scripts and explains fixes using DevPrep AI standards.

Prototype markup often lacks reusable structure, types, and accessibility. This skill analyzes the design and produces components suited to the target frontend stack.

Slow applications lose users and make incidents harder to diagnose. This skill gives Claude, Codex, and Claude Code practical patterns for faster systems.

Creating consistent feature modules by hand is slow and error-prone. This skill generates DevPrep AI module structure, starter TypeScript files, components, and quality checks.

Edge platforms differ in runtime limits, storage, and deployment behavior. This skill provides practical patterns for routing, authentication, caching, streaming, and rate limiting.

Hand-converting prototypes creates repetitive mapping and setup work. This skill analyzes HTML, guides component choices, scaffolds React files, and checks local conventions.

Inconsistent reviews miss defects and create unclear feedback. This playbook provides structured checklists, conventional comments, templates, and review workflows for reliable decisions.

Developers can lose time deciding where code belongs in DevPrep AI. This skill maps the structure and applies documented placement rules.

Inconsistent API contracts create integration errors and expensive migrations. This skill applies reusable REST, GraphQL, gRPC, OpenAPI, and AsyncAPI patterns to create clear contracts.

Teams need reliable patterns for adding LLMs, retrieval, tools, and agents to production software. This skill provides practical templates and references for RAG, vector search, function calling, observability, and cost control.