skills/cursor-codex-integration/SKILL.md
Use when configuring Cursor IDE or OpenAI Codex for game development — .cursorrules files, prompt conventions, context injection. Triggers: cursor, .cursorrules, codex, IDE setup.
npx skillsauth add fcsouza/agent-skills infrastructure-cursor-codex-integrationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Configuration files and prompt conventions for using Cursor IDE and OpenAI Codex with this game development skill ecosystem.
Trigger: cursor, codex, .cursorrules, cursor rules, AI IDE, codex instructions, prompt conventions, context injection, AI coding setup
claude-code-game-workflow — understand the skill ecosystem firstThe .cursorrules file sits at the repo root and is loaded automatically by Cursor. Use it to enforce the game-dev stack. Key rules to include:
# TypeScript strict mode — never `any`
# Runtime: Bun only — no npm/npx/yarn
# Backend: Elysia — never Express, Hapi, or Fastify
# Database: Drizzle ORM — no raw SQL, no Prisma, no TypeORM
# Genre-agnostic: shared engine code must never assume RPG/MMO/FPS
# Server-authoritative: all game logic on server, client is display only
# Narrative: quests/characters/lore require quest-narrative-coherence check FIRST
See templates/game-dev.cursorrules for the complete file with all ~25 rules.
How to use Claude Code with this ecosystem:
@skills/game-dev/engineering/postgres-game-schema/SKILL.md implement the inventory schema
claude-code-game-workflow/templates/game-project-claude.md. This file tells Claude Code which skills exist, when to use them, and project-specific constraints.claude-code-game-workflow/templates/claude-hooks-config.json for hooks that trigger the right skill based on file patterns (e.g., editing db/schema.ts triggers postgres-game-schema)."@engineering/postgres-game-schema/SKILL.md implement the inventory schema"
Claude Code will read the skill file and apply its patterns to the implementation.Five concrete prompt templates for common game dev tasks:
1. New feature:
Read @skills/game-dev/engineering/[relevant-skill]/SKILL.md. Then implement [feature] following the patterns there. Stack: Elysia + Drizzle + Neon. Use Bun runtime. No raw SQL.
2. Narrative content (quests/characters/lore):
Read @skills/game-dev/narrative/quest-narrative-coherence/SKILL.md first. Then create [quest/character]. Follow the 5-step coherence check and output a coherence report before writing any content.
3. Schema change:
Read @skills/game-dev/engineering/postgres-game-schema/SKILL.md. Add [table] following the JSONB + indexes + soft-delete patterns. Include the Drizzle migration command.
4. Bug fix:
Read the game-state-sync boilerplate. The bug is [describe]. Apply a server-authoritative fix only — do not move any game logic to the client.
5. Code review:
Review this [file] against the patterns in @skills/game-dev/engineering/[relevant]/SKILL.md. Flag violations of: server-authoritative architecture, genre-agnostic constraints, TypeScript strict type safety, and Drizzle ORM usage.
The same principles apply across AI IDEs:
.cursorrules content works in .github/copilot-instructions.md. Copy the rules file there and Copilot will apply them across the repo..windsurfrules at the repo root with identical syntax to .cursorrules. Use the same templates/game-dev.cursorrules content.claude-code-game-workflow — ecosystem navigationtools
Use when implementing client-server state synchronization, delta compression, optimistic updates, rollback netcode, or real-time game state reconciliation. Triggers: state sync, netcode, delta, rollback, interpolation, prediction.
testing
Use when designing virtual economies, currencies, sink/faucet balance, loot tables, crafting systems, shops, or inflation control. Triggers: economy, currency, sinks, loot, inflation, crafting, shop.
development
Audits existing game code against design principles — checks server-authority, schema conventions, auth security, payment safety, narrative coherence, and MVP scope drift. Extract the optional component name or path from the user's message (defaults to entire src/). Use after building components or before committing.
testing
Designs a single quest end-to-end — coherence check, objective tree, quest brief, and registry entry. Extract the quest name from the user's message. Requires docs/world-lore.md and docs/quest-registry.md.