Index of every page on this site: llms.txt
All skills
Run it yourself

Builds playable homage games (original takes on well-known games, with original names and art) whose in-game captures must hold up side by side with real shipped-game screenshots. Orchestrates separate planner, builder, and critic subagents through a brief with locked reference shots, a plan, a playable loop, a staged art pipeline, and live captures, then runs fresh-critic rounds against a per-game Visual Bar until WIN, with a diagnoser for stuck loops and a final orchestrator visual read before handoff. Use when asked to build, polish, or compare homage games or vertical slices where visual quality next to real games matters, when a game plays fine but its screenshots look like a browser toy beside the references, or when a reviewer keeps soft-passing weak visuals. Not for general app scaffolds or gameplay-only prototypes.

Use this Skill: https://skilld.dev/gh/ericzakariasson/skills/game-builder

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

visual-bar.md

≈1.7k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Visual Bar (hard gate)

The bar is per game, not a fixed global list. Locked refs are usually shipped AAA or high-end captures, and the game must look like it belongs in the same SxS frame. Judge every locked criterion on the current capture set at full resolution, then on the blurred SxS. Every criterion is PASS or FAIL. One failing still fails the criterion. WIN requires every criterion in art/BAR.md to PASS.

Contents

  • Building the bar (Stage A)
  • Universal core
  • Game-specific criteria
  • Automatic FAIL (visual)
  • Automatic FAIL (process)
  • Banned phrases

Building the bar (Stage A)

  1. Start from the universal core below. It is always present.
  2. Add game-specific criteria from BRIEF.md, the locked refs, and art/LOOK.md: what would make a player or reviewer reject this genre's still next to the refs. Prefer criteria you can see in the refs.
  3. Write the full list to art/BAR.md before Stage B. Each row: id, name, PASS when, FAIL signs, justifying ref ids, and whether it is core or game.
  4. Size band: minimum 5, maximum 15 criteria. The universal core is five. Add game-specific criteria until the refs are covered, without exceeding 15. A bar with only the five core criteria is allowed when the refs do not justify more. Fewer than 5 or more than 15 is a process FAIL.
  5. Freeze after Stage A. Mid-loop you may only BAR-EXPAND (add criteria) with a logged reason, and only while staying at or under 15. Never remove, merge away, or soften a criterion to exit. Never invent a weaker bar after FAILs.

Universal core

These five are in every art/BAR.md. Wording may adapt for 2D when the locked refs are themselves 2D, but the intent stays.

C1. Materials and lighting. PASS when large surfaces show material response (roughness or specular variation, normal or displacement detail, color variation); one readable key light with cast shadows; contact darkening at bases; exposure and grade inside the matched ref's palette; atmosphere consistent with the matched ref. FAIL: flat or unlit surfaces, fullbright, no shadows or contact darkening, toy-plastic palette.

C2. Geometry and silhouette density. PASS when matched crops show terrain, prop, and unit breakup comparable to the refs; readable sub-shapes; silhouette count in the same range; no primitive standing in for a real object. FAIL: stamped boxes, empty planes, tile seams, sparse where the ref is dense.

C3. SxS blur test. Blur both halves equally until text is unreadable, then downscale. PASS when value structure, palette, density, and light direction read as the same genre and production tier in every pair. FAIL when the game half reads flatter, emptier, or like a web page next to a shipped game.

C4. Live capture provenance. PASS when every still and walkthrough frame is from the running build at the delivered commit, at native resolution, with MANIFEST.md complete. Only cropping and labeling are allowed after capture. FAIL: mockups, composited plates, retouching, wrong build.

C5. Walkthrough consistency. PASS when sampled frames match still quality: no placeholders, no missing textures, no flicker, popping, or z-fighting, steady pacing, relevant UI states shown. FAIL: any of those in motion.

Game-specific criteria

Pick from what the locked refs actually show, and write PASS when and FAIL signs observables into art/BAR.md. These are examples, not a checklist to paste blindly:

  • HUD or diegetic UI craft, when refs show designed chrome (fighters, MOBAs, FPS, survival). Skip or narrow it when refs are HUD-light.
  • Hero or unit silhouette readability (fighters, MOBAs, stealth predators).
  • Cockpit or interior volume (space sims, racing cockpits).
  • Far-field or lane density (MOBAs, open battlefields).
  • Night grade or threat lighting (horror, night defense).
  • Readable gaze, cones, or threat volumes (stealth).
  • VFX that lights the scene (fighters, abilities, weapons).
  • Wet surfaces and reflections (harbors, rain).
  • Crowd or horde density (base defense, RTS).
  • Genre signature named from LOOK.md (whatever the refs share that a generic bar would miss).

Do not add vague criteria ("feels premium"). Every game-specific criterion needs a PASS when and FAIL signs pair and at least one justifying ref.

Automatic FAIL (visual)

Never excuse any of these when they apply to the locked bar and refs:

  • Any placeholder primitive standing in for a real object in any still or walkthrough frame.
  • A flat single-color surface dominating a still where the matched ref shows texture.
  • No cast shadows or no contact darkening where the refs have them.
  • Default fonts or unstyled form controls on any UI criterion that is in art/BAR.md.
  • A 2D plate as backdrop where the ref shows 3D space (unless the locked refs are themselves 2D).
  • Missing textures, z-fighting, flicker, or popping in the walkthrough.
  • Any SxS pair that fails the blur test.
  • Stills that are not from the live build at the delivered commit.

Automatic FAIL (process)

The Orchestrator fixes these before any score counts:

  • Refs not locked: no SOURCES.md, fewer than four refs, or refs swapped for weaker ones.
  • art/BAR.md missing, with fewer than 5 or more than 15 criteria, or with criteria that lack PASS and FAIL observables.
  • Any art/LEDGER.md row still placeholder at Stage D.
  • Fewer stills than locked refs, or SxS pairs not matched by scene and camera.
  • Missing manifest, missing walkthrough, or walkthrough under spec.
  • A Critic verdict that scores a different criterion set than art/BAR.md, contains a banned phrase, uses numeric scores, or cites fewer stills than delivered.
  • Model identifiers or credentials in any artifact.
  • Ref imagery found inside the game's asset or build directories.

Banned phrases

Any of these in a verdict invalidates it. Any of these in a Builder handback or Orchestrator status line is a process FAIL. Rewordings with the same meaning count.

  • "fine for a browser toy", "fine for a homage", "fine for a slice", "fine for a prototype"
  • "not photoreal, but", "not a AAA remake", "homage, not a remake"
  • "stylized low-poly is a valid choice" (when the locked refs are not low-poly stylized)
  • "close enough", "good enough", "mostly there", "acceptable for now", "nearly parity"
  • "WIN with reservations", "conditional WIN", "soft WIN", "provisional pass", "PASS with notes"
  • "given time constraints", "for this round", "considering the stack", "impressive for a web build"
  • "gameplay is solid, so", "big improvement over last round" used as grounds for PASS

If the stack cannot reach the bar with current art, change the stack or the art pipeline (see the escalation ladder in builder-protocol.md). Never lower the bar. Never delete criteria from art/BAR.md to make WIN easier.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning23d3 checks · Risk SAFE
  • Gen Agent Trust Hub23d

    The 'game-builder' skill is a sophisticated orchestration framework for creating high-quality games through a multi-agent workflow (Planner, Builder, Critic). It emphasizes visual parity with reference images and enforces strict procedural gates. Security analysis shows the skill incorporates industry best practices, such as automated credential scanning, role separation to prevent prompt leakage, and strict requirements for fresh sessions. It utilizes a companion skill from the author's own repository, representing standard vendor functionality without introducing external risks.

  • Socket23d

    No alerts

  • Snyk23d

    Risk: MEDIUM · 1 issue

Signed by skilld at 090915f. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Activeupdated 3 weeks ago
Other metadata
compatibility
Needs a host that can run roles as separate subagents or fresh sessions, and a critic that can view full-resolution images.

README badge

README badge for ericzakariasson/skills/game-builder