agentleFS
fcsouzafcsouza/agent-skills/skills/game-dev/SKILL.md

game-dev

Skill26 starsChanged yesterdayMITView on GitHub

Entry point for game development work: routes a task to the right game-dev skill and states the project conventions. Use when starting or continuing work on a game project, or when it is unclear which game-dev skill applies.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run all game logic on the server and treat the client as display only.

    line 91
  2. Very specific:

    Read postgres-game-schema before writing database code.

    line 93
  3. Distinctive:

    Keep shared code free of genre-specific mechanics.

    line 94
  4. Distinctive:

    If a skill in the table is not installed, run:

    line 98

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Game Dev

Find the task in the table, then load the skill named beside it. Commands run a whole workflow; start with them when one fits.

Route

Design

NeedSkill
Design loop, core loop, difficulty, rewardsgame-design-fundamentals
Currencies, sinks and faucets, loot, crafting, shopsgame-economy-design
Translation: string IDs, plurals, text expansion, pseudo-localizationgame-localization
Level layout, pacing, encounter placement, difficulty curvelevel-design
Persistent world with many players: shards, economy at scale, alts, roles, guilds, endgame, live opsmmo-design
Dungeons, maps, terrain, seedsprocedural-gen
Quests, missions, objectives, rewardsquest-mission-design
HUD, menus, onboarding, accessibilityui-ux-game

Art and assets

NeedSkill
Sound effects, music, voiceelevenlabs-sound-music
3D models from Blender: headless export, glTF and FBX, LODs, collision, bakinggame-blender

Engineering

NeedSkill
Login, sessions, roles, WebSocket authbetterauth-integration
Queues, delayed jobs, scheduled workbullmq-game-queues
Cheating: input validation, hidden state, dupes, detection, bansgame-anti-cheat
Game server, WebSocket, rooms, tick loopgame-backend-architecture
Browser client: stack choice (web, Phaser, PixiJS, Three.js), loop, input, assetsgame-client
Godot 4: GDScript or C#, headless export, WebSocket client to our servergame-godot
Saves, slots, save versions and migration, cloud sync conflictsgame-save-load
Lag, desync, jitter; prediction, reconciliation, interpolationgame-state-sync
Unity: Unity CLI and batch mode, API traps, WebSocket client to our servergame-unity
Player events, retention, funnelsgameplay-analytics
Pairing players, lobbies, ranksmatchmaking-system
Tables, player data, inventory, Drizzlepostgres-game-schema
Leaderboards, presence, rate limits, cache-asideredis-game-patterns
Skill trees, talent trees, unlock graphsskill-progression-trees
Payments, purchases, subscriptions, webhooksstripe-game-payments

Narrative

NeedSkill
Characters, NPCs, dialogue, arcscharacter-design-narrative
Check new content against existing lore and questsquest-narrative-coherence
Branching story, acts, player agencystory-structure-game
World, factions, geography, timelineworldbuilding

Infrastructure

NeedSkill
Pipelines, deploys, migrations, a database per pull requestci-cd-game
Logging, alerts, server health, tick time, error ratesmonitoring-game-ops

Workflows

NeedSkill
Start a project: interview, research, MVP plan/game-architect (game-architect)
Check economy, difficulty or progression balance/game-balance (game-balance)
Build one component from the plan/game-build (game-build)
Add a post-MVP feature to the plan/game-expand (game-expand)
Add a faction, location, NPC or event to the lore/game-lore (game-lore)
Design one quest end to end/game-quest (game-quest)
Audit code against the design rules/game-review (game-review)

Installed as the Claude Code plugin, the commands are /game-dev:<name>, for example /game-dev:game-build.

Project files

  • docs/mvp-first-draft.md: the MVP plan, created by game-architect, then extended by game-expand.
  • docs/build-registry.md: components built so far, created and kept by game-build.
  • docs/world-lore.md: canonical world state, created by game-architect, then kept by game-lore.
  • docs/quest-registry.md: every quest, created by game-architect (or by game-quest when missing), then kept by game-quest.

Rules

  • Run all game logic on the server and treat the client as display only. A client the player controls can be edited.
  • Apply quest-narrative-coherence before writing a quest, character or lore entry. New content then cannot contradict what exists.
  • Read postgres-game-schema before writing database code. Queries written without the schema drift from it.
  • Keep shared code free of genre-specific mechanics. Genre-specific logic in shared code breaks the next game that reuses it.

Missing skill

If a skill in the table is not installed, run:

bunx skills add fcsouza/agent-skills --skill <name>

New project

Paste templates/game-project-claude.md into the game repo as its CLAUDE.md and fill in the placeholders.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.