gigio1023/game-studio · Archived

game-direction

Use when game work starts without a stable direction: a creator asking "what game should I make", turning taste or references into a concept and creative brief, adopting an existing direction or pitch, judging "is this idea worth building" or whether to continue or pivot or kill an existing project, reviewing whether direction artifacts are stable enough for production, or auditing an ambiguous game repo, prototype, or notes. NOT for milestone planning or implementation (use game-production) an…

First seen Jul 10, 2026

Installation

$ npx skills add gigio1023/game-studio --skill game-direction

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from gigio1023/game-studio.

npx skills add gigio1023/game-studio

Browse all from gigio1023/game-studio

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch main
Open issues 0
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 4,475 B
  • docs SUMMARY.md 586 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 5 installs

SKILL.md

Game Direction

Turn creator intent into a direction the creator recognizes as theirs: a short creative brief plus one named first playable build. Direction becomes stable through play, not through longer prose, so every route here ends with a playable build target, never with documentation alone.

Route First

Classify the entry before doing direction work:

Mode When First move
direction-carry-in A thesis, pitch, mandate, prototype, or public promise already exists Preserve it. Audit what is still unproven. Do not rewrite direction because an artifact is incomplete.
taste-to-game The creator has taste, references, desired feelings, or hard noes, but no stable direction Elicit taste, then build a small concept slate. Read references/taste-to-game.md.
brownfield-unknown A repo, notes, or assets exist but authority and direction are unclear Audit before writing anything new: separate playable truth from claims in notes, record who owns direction, then reroute to one of the modes above.

Two deeper tools:

  • When the creator needs a real conversation — philosophy, player role,

whole-game frame, first playable build — run the interview in references/creator-interview.md.

  • When judging whether a thesis, pillars, player-experience target, or concept

slate is solid, use the checklists in references/direction-review.md.

Authority

The human owns taste, concept selection, Direction Lock (the explicit decision that fixes direction), and public promises. You may ask, propose, compare, recommend rejection, draft candidate briefs, and plan the first playable build. When the next step needs one of the human-owned calls, stop and return USERDECISIONNEEDED with the exact decision required — do not choose for them, and do not present your own inference as their taste.

An audit or review request authorizes inspection and findings, not rewriting project files or locking direction. A request to create or update direction artifacts authorizes reversible local drafts; keep them candidate material until the creator accepts them. Publishing a promise or making a destructive change still requires explicit approval.

Concept Status

Track where the project is with one status: UNROUTED, TASTEDISCOVERY, CONCEPTSLATE, CREATIVEBRIEF, FIRSTBUILDPLANNED, DIRECTIONLOCKED, or PARKED. A concept slate or candidate brief is not active scope until the creator selects it; parked and rejected concepts keep their reasons. Always report one exact value from this set; prose such as "unchanged" is not a status. Naming a build for an accepted carry-in can produce FIRSTBUILDPLANNED, but never implies Direction Lock.

Output

End a routing or discovery pass with:

Entry mode:
Direction source:
Authority owner:
Concept status:
Creative brief or candidate brief:
First playable build:
Safe assumptions made:
Human decision needed:
Next step:

Preserve every field; write none when one does not apply. Tag load-bearing facts by source: user fact, repo fact, safe assumption, or blocker. Do not close direction work while a blocker remains.

Gotchas

  • Do not ask a novice creator for genre, core loop, engine, or monetization

before taste has been elicited; ask for remembered moments and desired feelings first.

  • Do not generate idea catalogs. Three to five comparable, rejectable concepts

beat twenty.

  • Do not smooth away contradictions in the creator's taste too early; surface

one contradiction at a time.

  • Do not turn discovery into documentation work: if the session produces prose

but no named playable build, the route is not done.

  • If the first playable build cannot be named in one sentence (one player

action, one visible consequence), the concept is not ready for Direction Lock.