React apps can slow down because renders, lists, bundles, images, and heavy computations grow unnoticed. This skill gives practical patterns for diagnosing and reducing those costs.

React failures often hide behind silent returns and weak error messages. This skill gives Claude, Codex, and Claude Code practical patterns for logs, tracing, and safer diagnostics.

Async React bugs create race conditions, missed state updates, and unreliable effects. This skill guides reviews, fixes, and lint rules for safer async code.

Async Python tests can fail silently, leak database state, or miss awaited behavior. This skill gives pytest-asyncio patterns for fixtures, mocks, HTTP clients, and full-flow tests.

SQLModel async projects often fail from lazy loading, schema drift, and session mistakes. This skill gives focused patterns for models, queries, relationships, and ORM debugging.

Pydantic v1 habits can cause validation and serialization defects. This skill provides Pydantic v2 patterns for schemas, validators, configuration, unions, and serialization.

FastAPI projects often drift into inconsistent routes, responses, and schemas. This skill gives practical patterns for endpoints, validation, errors, middleware, and OpenAPI.

Alembic autogenerate can create migrations that lose data or lock tables. This skill gives practical PostgreSQL patterns for safer schema changes, reviews, testing, and production rollout.

Async Python code can leak sessions, block responses, and deadlock under load. This skill gives FastAPI and SQLAlchemy patterns for safer async database work.

React interfaces often fail to show clear loading, error, empty, and submission states. This skill gives practical patterns for predictable async UI behavior.

GraphQL client work often creates duplicate operations, weak error handling, and stale generated types. This skill guides consistent .gql files, Apollo generated hooks, codegen steps, and mutation UI safeguards.

Formik forms often repeat validation, error display, and submit-state logic. This skill provides concise patterns for reusable fields, Yup schemas, GraphQL submissions, edit forms, and multi-step flows.

Inconsistent UI components slow product work and create visual drift. This skill guides Claude, Codex, and Claude Code to use shared tokens, layouts, and component patterns.

Large repositories make symbol lookup slow and repetitive. This skill adds repo-map MCP tools that index project symbols and retrieve relevant code context quickly.

AI coding sessions can skip requirements, design, or testing when pressure rises. This skill keeps work inside a strict waterfall flow with documented gates.

Inconsistent commit messages make reviews and release notes harder to scan. This skill formats concise Conventional Commit summaries with clear 72-column bodies.

Remix apps need cache freshness without fragile invalidation logic. This skill guides Redis-backed caches, SSE revalidation, TTL choices, and tests.

Complex agent work can fail when one context window carries every task. This skill helps Claude, Codex, and Claude Code design agent systems with clear handoffs.

Long conversations and large prompts can make AI agents lose important information or follow conflicting context. This skill helps Claude, Codex, and Claude Code identify degradation patterns and choose mitigation strategies.

Code reviews can become inconsistent, vague, or overly focused on style. This skill applies a three-pass method, practical checks, and focused feedback.