graphql-schema-generation

name	graphql-schema-generation
description	Generate a GraphQL schema for a domain, including types, queries, mutations, subscriptions, directives, and resolvers. Do NOT use for REST API design, OpenAPI spec generation, or GraphQL client code generation.
version	1.1.0
time_saved	Manual: 3–6 hours | With skill: 20–40 minutes
license	Proprietary — Personal Use Only
category	coding
complexity	Advanced
tokens	~4000
tags	["graphql","schema","types","resolvers","mutations","subscriptions","api-design"]
author	vheins

Skill: GraphQL Schema Generation

Purpose

Generate complete, best-practice GraphQL schemas including SDL and resolver outlines.

Input

Variable	Type	Req	Description
tech_stack	string	Yes	e.g., "Node.js + Apollo Server"
domain_description	string	Yes	Entities, relations, and business rules
operations	string	Yes	Required queries, mutations, subscriptions

Instructions

• Types: Define Object types (nullability), Input types (for mutations), Enums, and Unions. Use Connection types for pagination (Relay spec).
• Queries: Implement single lookup, paginated lists, and nested relations using descriptive names.
• Mutations: Return mutated payload objects with result/error unions. Add validation rules as comments.
• Subscriptions: Define triggers and filter arguments for real-time updates.
• Resolvers: Outline structure, identify N+1 risks (DataLoader), and define auth check placement.

Edge Cases

Case	Strategy
Circular refs	Use lazy types; enforce resolver depth limits.
Large results	Enforce first arguments with maximum caps.
Field-level Auth	Apply @auth directives; specify field-level resolver placement.

Schema Flow

flowchart TD
    A([Start: GraphQL Schema Generation]) --> B{Schema source?}
    B -- Domain model --> C[Map entities → GraphQL types]
    B -- REST API --> D[Convert endpoints → queries/mutations]
    B -- Database schema --> E[Introspect tables → types + resolvers]
    C & D & E --> F[Define scalar types: ID, String, Int, DateTime, custom]
    F --> G[Model relationships: one-to-one, one-to-many, many-to-many]
    G --> H{Nullable vs non-null?}
    H --> I[Apply ! where field is always present]
    I --> J[Define Query type: list, single, filtered, paginated]
    J --> K{Mutations needed?}
    K -- Yes --> L[Define Mutation type: create, update, delete, batch]
    K -- No --> M
    L --> M{Subscriptions needed?}
    M -- Yes --> N[Define Subscription type for real-time events]
    M -- No --> O
    N & O --> P[Define input types for mutations]
    P --> Q{Auth/access control?}
    Q -- Yes --> R[Add directives: @auth, @deprecated, @skip, custom]
    Q -- No --> S
    R & S --> T{Pagination strategy?}
    T -- Cursor-based --> U[Add Connection/Edge/PageInfo types]
    T -- Offset-based --> V[Add limit/offset args]
    U & V --> W[Validate schema: no circular deps, no orphan types]
    W --> X[Generate SDL file + resolver stubs]
    X --> Y([Output: GraphQL schema + resolver scaffolding])

Examples

• Input Example
• Output Example

Quality Gate

1. Are inputs separate from objects?
2. Is pagination enforced on lists?
3. Are error unions included in mutations?
4. is nullability explicitly handled?
5. is the schema Relay-compliant?

MCP Dependencies

• @upstash/context7-mcp: Library documentation and examples.

Changelog

Version	Date	Description
1.1.0	2026-03-20	Restructured: moved examples/references, added compatibility/license
1.0.0	2026-03-20	Initial release