practices▌

Tamagui v1.x patterns beyond fundamentals: Config v4, compiler optimization, compound components, and gotchas.

Node.js architecture and framework selection guidance based on deployment context and project requirements. \n \n Framework decision tree covers Hono (edge/serverless), Fastify (performance), NestJS (enterprise), and Express (legacy), with selection criteria tied to cold start time, ecosystem size, and team experience \n Covers layered architecture principles (controller, service, repository), async patterns (async/await vs Promise.all/allSettled), and centralized error handling with appropriate

Email verification, password reset flows, and customizable password policies for Better Auth. \n \n Supports email verification with optional enforcement to block sign-in until verified, plus configurable token expiration and single-use reset tokens \n Password reset flows with built-in security: background email sending, timing attack prevention, dummy operations on invalid requests, and optional session revocation on reset \n Configurable password length limits (default 8–256 characters) and c

Standard flake.nix structure:

28 ClickHouse best practices rules organized by schema design, query optimization, and data ingestion strategy. \n \n Covers three critical areas: primary key and data type selection (immutable design decisions), JOIN and query optimization, and insert batching with mutation avoidance \n Includes 28 rules prioritized by impact, with schema design and query optimization rules marked CRITICAL due to ClickHouse's columnar storage and sparse index mechanics \n Provides structured review procedures f

When running Playwright tests from Claude Code or any CLI agent, always use minimal reporters to prevent verbose output from consuming the context window.

Architectural patterns for coordinating TanStack Query, Router, and Start across full-stack applications. \n \n Covers setup, data flow, caching, and SSR integration with 13 rules organized by priority and category \n Emphasizes loader-based data fetching with ensureQueryData, suspense queries in components, and server functions as query sources \n Provides unified caching and invalidation strategies to prevent duplication and ensure single source of truth \n Includes automatic SSR dehydration/h

Simple, pragmatic, opinionated. Only what matters for writing production-grade SQLAlchemy and Alembic code.

$22

Use this skill when: