graphql-architectlisted

# GraphQL Architect Design GraphQL APIs that are explicit, stable, performant, and easy for clients to consume. ## Use This Skill For - New GraphQL schemas or modules - Resolver and mutation design - Pagination, filtering, and search - N+1 reduction and batching - Authorization, validation, and error-shaping for GraphQL ## Workflow ### 1. Read the Existing GraphQL Shape - Inspect schema files, resolver structure, and context setup - Find at least 3 existing resolver or schema patterns before adding new ones - Match the dominant conventions for naming, nullability, and error handling ### 2. Design the Contract Before Implementation - Start from client use cases, not database tables - Define the schema shape first: queries, mutations, input types, output types - Keep types stable and explicit; avoid ambiguous fields or overloaded mutations ### 3. Implement Resolver Boundaries Cleanly - Keep resolvers thin; push business logic into services - Validate inputs at the API boundary - Enforce authorization close to the entry point - Avoid leaking database details directly into schema types ### 4. Control Performance Early - Use batching/data loaders where related fields fan out - Prefer cursor pagination over ad hoc offset pagination for large lists - Avoid fields that trigger hidden expensive work without clear client intent - Measure N+1 risk any time a list returns nested entities ### 5. Verify Client Ergonomics - Field names should describe domain concepts, not stor