Game-Creation-Agents: Autonomous Game Production

Tencent Multimodal RL Team; Xiangxin Zhou; Jiaqi Tang; Ke Ma; Wenhao Yang; Jiarui Yao; Boye Niu; Haiyang Guo

cs.AIOct 12, 2026

Game-Creation-Agents: Autonomous Game Production

Authors: Tencent Multimodal RL Team, Xiangxin Zhou, Jiaqi Tang, Ke Ma, Wenhao Yang, Jiarui Yao, Boye Niu, Haiyang Guo,

Organizations: Multimodal Reinforcement Learning Team, Tencent

Abstract

Game-Creation-Agents (GCA) turns a natural-language request into a runnable Phaser 4 browser game. The pipeline first builds a multimodal game design document that couples mechanics and scene contracts to grounded visual references. It then projects that document into a runtime scaffold, machine-readable contracts, and deterministic tests. Eight specialized agents next carry out production: scenes are implemented serially, assets are produced in parallel, and one reduced manifest joins the two branches into a single build. Build-linked real-UI checks and a code-blind critic then verify that build and decide whether the game may ship. Each blocker returns to the agent that owns the affected artifact for a bounded repair, and the game is released once all the gates pass. We evaluate finished games separately, through their rendered interfaces alone, over prompt-matched tasks run under a common protocol. GCA achieves stronger requirement satisfaction, core mechanics, content depth, and functional visuals than a comparable open pipeline. In a side-by-side human study, raters also score GCA's games higher on every dimension and prefer them overall.

Figures & tables

System	Primary scope	Executed or rendered evidence	Role of the outcome
GameGPT ( Chen et al., 2023 )	Multi-stage game development	Task execution plus role-specific review	Workflow review and completion
OpenGame ( Jiang et al., 2026b )	End-to-end Phaser game and asset creation	Build/test verification; companion benchmark executes the game	Iterative debugging; separately reported benchmark scores
MarioGPT ( Sudhakaran et al., 2023 )	Text-conditioned Mario-level generation	Generated-level sampling and novelty search	Content generation and diversity
GameDevBench ( Chi et al., 2026 )	Godot development tasks	Engine validation; screenshots and runtime video studied as agent feedback	Development-agent benchmark
OpenGame-Bench ( Jiang et al., 2026b )	Finished web games	Headless browser execution and multimodal judging	Build Health, Visual Usability, and Intent Alignment scores
GameCraft-Bench ( Luo et al., 2026 )	Complete Godot games and replay traces	Replayed inputs, gameplay video and frames, hidden rubric	Core Mechanics, Content Depth, Functional Visuals, and Art scores

Table 1: Representative game systems and the role of evidence.

Agent	Responsibility	Owned artifact class
design	Author the multimodal GDD: the design specification plus linked concept mockups per screen and grounded crops of each element	Design and concept evidence
level-design	Materialize or revise authored levels; dispatched only when routing calls for them	Authored level set
scaffold	Project the multimodal GDD into a Phaser 4 shell plus the machine-readable contracts and deterministic tests that later gates read	Runtime shell, contracts, and tests
implement	Implement one scene’s gameplay and its feature tests; dispatched once per scene, serially	Scene implementations and feature tests
assets	Generate and register visual assets for one batch through the control plane, with per-asset metadata sidecars	Visual assets and metadata
asset-integration	Wire the finalized manifest into runtime preloads, animations, and display parameters, then seal the integration receipt	Runtime asset wiring and sealed receipt

Table 2: Specialized agents and their owned artifacts.

Figure 3: Three Deckbuilder scene contracts link multimodal GDD concepts to matching downstream runtime captures.

Figure 4: Animated-sprite production alternates bounded generation, checks, retries, and one post-join finalization.

Figure 5: One verification epoch. Checks run against the freshly sealed build and produce the evidence shown here, from a Minesweeper run. The gate and the critic then read the sealed report and either advance the run or return a blocker to its owner for a new epoch.

Category	Tools	Scope, with a representative tool
Lifecycle	9	Workspace leasing, preflight readiness, scaffold reproject receipts, and the transactional state journal — write_pipeline_state
Asset generation and validation	20	Image and video synthesis, matting, spritesheet packing, and the closed-loop producers that retry against VLM rubrics — produce_validated_sprite_action
Verification	16	The checkpointed verification DAG, its individual producers, and the gate runner — run_verify_dag
Concept grounding	3	Detect element boxes on an approved mockup and crop them into art seeds — detect_layout
Post-processing and matting	3	CPU compositing plus the two matting backends, remote and deterministic — rmbg2_matting
Video understanding and recording	3	Record a running game with sampled engine state, then analyze only that sealed recording — record_with_state

Table 3: MCP control-plane tools by category.

Figure 6: Single-writer mutation control accepts owned writes and routes rejected cross-component changes to the declared owner.

No.	Task	Family	No.	Task	Family
1	Minesweeper	Puzzle	11	Blackjack	Card
2	Snake	Arcade	12	Endless Runner	Runner
3	Tap Flyer	Arcade	13	Fruit Slicer	Action
4	Shmup	Action	14	Solitaire (Klondike)	Card
5	Match-3	Puzzle	15	Link Match	Puzzle
6	Tower Defense	Strategy	16	Gold Miner	Arcade

Table 4: Fixed 20-task paired corpus with game families, numbered in report-local sequence. Tasks 1–10 are listed on the left and 11–20 on the right. Task names describe gameplay genres and mechanics; they do not refer to any commercial game.

Metric	Output	Rubric question	Scoring
Requirement Satisfaction	$[0,1]$	What fraction of the observable prompt MUST requirements is satisfied?	Mean of all prompt-derived Core Mechanics, Content Depth, and Functional Visuals items.
Core Mechanics	$[0,1]$	Do controls, rules, state transitions, outcomes, and restart behavior work?	Weighted fraction of passed deterministic MUST-item checks.
Content Depth	$[0,1]$	Are requested modes, levels, progression, and terminal content present?	Weighted fraction of passed deterministic MUST-item checks.
Functional Visuals	$[0,1]$	Are required game state, HUD, action feedback, and outcomes visible?	Mean of direct evidence-grounded GPT-5.6 MUST-item scores.
Art & Presentation	$[0,1]$	Is the presentation coherent, readable, and visibly finished?	Mean of three shared GPT-5.6 visual-quality items.

Table 5: Test metrics and scoring rubric.

	Mean $\pm$ standard deviation
Metric	GCA	OpenGame
Requirement Satisfaction	0.786 $\pm$ 0.151	0.656 $\pm$ 0.247
Core Mechanics	0.852 $\pm$ 0.144	0.655 $\pm$ 0.267
Content Depth	0.712 $\pm$ 0.286	0.651 $\pm$ 0.267
Functional Visuals	0.711 $\pm$ 0.192	0.667 $\pm$ 0.248
Art & Presentation	0.797 $\pm$ 0.099	0.827 $\pm$ 0.068

Table 6: Aggregate prompt-grounded results. Each system contributes one game per task and every game is scored three times, so a cell averages 60 evaluation runs over 20 games and its deviation spans tasks as well as scoring repeats. Bold marks the higher mean in each row. Core Mechanics and Content Depth derive from deterministic checks, Functional Visuals and Art & Presentation from grounded GPT-5.6 judgments, and Requirement Satisfaction averages the prompt-derived items across the first three dimensions. Null values are excluded, not converted to zero.

Reported quantity	GCA	OpenGame
Rubric ratings on a 1–5 scale (mean $\pm$ standard deviation)
Requirement Satisfaction	3.55 $\pm$ 1.36	3.28 $\pm$ 1.47
Core Mechanics	3.40 $\pm$ 1.31	3.16 $\pm$ 1.36
Content Depth	3.43 $\pm$ 1.36	3.28 $\pm$ 1.40
Functional Visuals	3.33 $\pm$ 1.35	3.29 $\pm$ 1.33
Art & Presentation	3.57 $\pm$ 1.13	3.03 $\pm$ 1.19

Table 7: Human study on 20 prompt pairs. Ten participants each rated all 20 pairs, giving 200 participant–game ratings per system. Bold marks the higher value in each row, and Overall is the mean of the five dimensions. Of the 200 preference choices 59 were “no preference”; excluding ties, GCA is preferred in 86 of 141 decisive choices (two-sided sign test,

p=0.011

Figure 7: Gold Miner formal-run evidence: retrieval, exact $500 credit, shop purchase, and level-2 transition in one continuous recording.

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Tool	W	Purpose
ping		Health check: workspace and credential status.
init_workspace		Open one game directory under explicit create or resume intent; pins the inode and takes an exclusive repository-wide lease.
close_workspace		Detach the workspace and release its lease; gated tools then fail.
list_tools_summary		One-line summary of every registered tool.
preflight_check		Startup readiness: credentials, R2 upload config, filesystem, Playwright, ffmpeg/ffprobe, matting endpoint.
asset_pipeline_smoke_test	✓	End-to-end probe before asset work: tiny image upload, matting, video credentials.

Appendix

Table 8: Lifecycle tools.

Tool	W	Purpose
generate_image	✓	Text-to-image; wraps the prompt in a chroma-key background constraint for mattable asset types.
edit_image	✓	Reference-conditioned edit; derives a game-ready sprite from an approved concept crop.
generate_seedance_video	✓	Image-to-video or text-to-video on the active provider channel.
extend_background	✓	Seamless rightward outpainting for static backgrounds.
build_i2v_prompt		Wrap an animation prompt in the validated constraint block: background lock, hard-edged effects, no glow or fade.
rmbg2_video_matting	✓	Per-frame matting to an RGBA sequence plus encoded transparent outputs.

Appendix

Table 9: Asset-generation and validation tools.

Tool	W	Purpose
run_verify_dag	✓	Canonical pass: opens or resumes an epoch, runs producers concurrently, fans in, finalizes in a finally block, and gates.
begin_verify_run	✓	Low-level epoch open under the workspace lock; diagnostics and recovery only.
run_tests	✓	Playwright suite against the epoch URL; a runtime token binds the browser to this workspace.
play_game	✓	Closed-loop play agent scoring itself against the executable objectives specification.
visual_diff	✓	Whole-screen and per-element perceptual comparison against concept mockups and crops.
capture_layout	✓	Deterministic capture of rendered object bounds for the layout axis.

Appendix

Table 10: Verification tools.

Tool	W	Purpose
Concept grounding (3)
detect_layout		Precision vision pass: tight canvas-relative boxes with roles and anchors for a mockup’s elements.
ground_crop		Cut detected boxes into per-element reference images that seed asset generation.
fit_image_to_resolution		Fit a concept mockup to the exact game resolution; design files only, never runtime assets.
Post-processing and matting (3)
rmbg2_matting	✓	Remote soft-alpha background removal; the preferred backend.

Appendix

Table 11: Concept-grounding, post-processing, video, and layout tools.

Group	Checks
Structure and contracts	basic_structure , levels , level_runtime , test_contract , rubric_contract , asset_manifest , asset_runtime_references
Build completion	scenes_implemented , assets_generated , asset_integration
Execution evidence	tests_passing , test_coverage , verify_report , visual_verified
Alignment axes	visual_alignment , gameplay_alignment , physics_alignment , element_alignment , layout_alignment , visual_diff , scene_coverage
Asset integrity	bbox_coverage , crop_coverage , aspect_drift , matting_complete , no_fx_spritesheets , subject_margin , slot_validation
Genre and interaction	interaction_contract , ui_contract , ui_layout , tower_defence_contract

Appendix

Table 12: Checks in iteration_ready , grouped by function.

Figure 8: Composite gate sizes, from the seven-check scaffold boundary to the 39-check release gate.

Key	Value	Effect
Iteration budget and evidence freshness ( pipeline )
min_iterations / max_iterations	1 / 2	Required fix rounds, and the hard cap
max_consecutive_blocks	5	Safety valve ahead of the budget
min_screenshots / min_unique_screenshots	3 / 3	Distinct screenshots required as evidence
staleness_tolerance_seconds	60	Clock-skew slack on freshness
max_analysis_age_minutes	30	Analyses expire after this

Appendix

Table 13: Default pipeline, gate, provider, and verification thresholds.

ID	Name	Condition for passing
I1	Identity match	Species, gender, era, art style, and notable features match the description exactly; no substitutions or embellishments.
I2	Hand-prop attachment	Every prop shows a visibly gripping hand or forearm; a prop merely near the body fails, however plausibly placed.
I3	Limb integrity	All expected limbs present, torso-connected, at realistic joint angles; none missing, extra, duplicated, or clipping.
I4	Foot anchor	Lowest visible point in the bottom 30% of the canvas; a floating or vertically centered figure fails.
I5	In-frame	Character and props unclipped, with at least 20% margin above the head and 15% each side for later motion.
I6	Background pure	Uniform solid chroma-key color; no gradient, shading, noise, secondary object, or border.

Appendix

Table 14: Image rubrics I1–I8, applied to sprite first frames.

ID	Name	Condition for passing
Framing group, judged on the raw video
R1	In-frame	Nothing is clipped in any frame, including peak motion: every swing, sweep, dash, and trail stays fully inside the canvas. Feet on the bottom edge pass.
R2	Scale locked	Size stays within roughly 10–15% of the first frame, judged late-against-first so creep is caught. Extending a limb passes; scaling the body does not.
Content group, judged on the matted output
R3	Identity preservation	Same colors, patterns, face, and proportions in every frame; no drift, swap, or morphing.
R4	Hand-prop attachment	Held items stay anchored to a hand or the body; nothing floats, detaches, or hovers without a visible grip.

Appendix

Table 15: Video rubrics R1–R7.

Dimension	Scale, from 0.1 through 0.3 and 0.5 to 0.8+
visual_quality	Incoherent; rough or accidental; intentional but uneven; profile-coherent, polished
gameplay_mechanics	No gameplay evidence at all; partially working; mostly working; all designed mechanics visible
completeness	Multiple major features missing; one or two missing; all present but rough; complete and refined
polish	No transitions or feedback; basic transitions; has juice; feels polished
asset_fitness	Missing, broken, or illegible; inconsistent or poorly integrated; runtime-fit but rough; cohesive and correctly integrated

Appendix

Table 16: Critic scoring dimensions and anchors.

Agent	Inputs	Returns
design	Required : game_idea , genre , design_axes , interaction_profile , quality_profile , level_model , game_dir , prompt_path	status , gdd_path , sections , scene_count , asset_count , concept_mockups , scene_coverage , ref_crops , warnings , issues
	Writes : the design document and multimodal evidence under design/ .
level-design	Dispatch : authored levels only Mode : create , or revise with structured issues Required : gdd_path , genre , level_model , game_dir	stage_count , level_files , has_boss , extensions , design_change_requests
scaffold	Mode : scaffold , or reproject with non-empty changed Required : gdd_path , game_dir , config_path , quality_profile , level_model	contracts , files_created , stale_owned_paths , stale_receipt_id
	Reproject deletion is receipt-mediated: scaffold seals its own stale list, and the orchestrator can only consume that receipt, never supply or widen the paths.
implement	Dispatch : one scene Required : scene_name , gdd_path , game_dir , genre , interaction_profile , quality_profile , level_model , url Optional : visual_evidence , on visual fix rounds only	scene , files_changed , tests , issues , design_change_requests

Appendix

Table 17: Phase-agent dispatch interfaces.

Field	Meaning
game_dir	Absolute game-project root; MCP tools take the same path under the same name
gdd_path	Absolute path to the game design document
genre	Startup-selected genre; downstream agents never re-route it
quality_profile	One of art_directed , responsive , pixel_art , minimal
level_model	One of authored , procedural , board , none
interaction_profile	Startup-selected physics and interaction route

Appendix

Table 18: Fields shared across dispatch interfaces.

Phase	Continuation rule
Design, level design, scaffold	No automatic retry. Design converts a bounded concept-generation failure into its documented warning fallback: complete layout and mechanics ship without the image.
Implement	No automatic scene retry; partial issues enter the fix loop, and resume targets only incomplete scenes.
Assets	Internal retries are bounded by configuration; a terminal partial may continue with only the missing keys.
Asset integration	One full dispatch first; scoped continuations are for confirmed timeouts only and run sequentially in the order config, preloads, animations, visual, verify.
Verify	Resume the same epoch from checkpoints after a confirmed interruption. Never restart completed nodes, and never open a fresh epoch for a transport failure. A new run starts only after the build changes or the previous one finalizes.
Critic	No automatic retry; a failed judgment blocks assessment rather than defaulting to either verdict.

Appendix

Table 19: Per-phase retry and continuation policy.

AI Coding Agents

AI coding agents use language models and software tools to plan, edit, test, and maintain code with varying levels of autonomy.

See all 321 papers

From Code to Playable Systems: CEGame-Bench for Long-Horizon Game Creation Agents
Oct 19, 2026Zixuan Dong, Yanghai Wang, Hanyan Bian +8Automated Game Testing Software Engineering Agents
National University of Defense Technology · Nanjing University · The University of Tokyo +2
GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?
Jun 16, 2026Tongxu Luo, Rongsheng Wang, Jiaxi Bi +22AI Agent Benchmarks Agentic Code Generation
1The Chinese University of Hong Kong, Shenzhen · 2Shenzhen Loop Area Institute · 4USTB +5
VibeGame: Prompt-to-Game Development with AI-Native Engine and Self-Evolving Adversarial Agent Team
Oct 22, 2026Wenbo Hu, Ken Li, Jiazhe Wei +13AI Coding Agents Multi-Agent LLM Systems
PRLab, Nanjing University · S-Lab, Nanyang Technological University

Oct 19, 2026J/K move · Enter open · S save

From Code to Playable Systems: CEGame-Bench for Long-Horizon Game Creation Agents

Zixuan Dong, Yanghai Wang, Hanyan Bian, Jiaxi Tao+7

National University of Defense Technology · Nanjing University · The University of Tokyo+2

Automated Game Testing Software Engineering Agents Long-Horizon Agent Evaluation

Open paper AskPDF arXiv

A game ships when a player can launch it, finish it, and return to a saved world, none of which a successful build guarantees, yet game-creation agents are mostly tested on one engine or on one-prompt browser games. We present CEGame-Bench, the first long-horizon benchmark for complete-game creation across five production engines: Defold, Godot, Roblox Studio, Unity, and Unreal Engine. Agents build each game from an empty project and a design document in runs averaging up to 295 steps and 65M tokens; the document doubles as a contract, and a hidden evaluator accepts a game only by real-input play. Under four scaffolds, agents earn most integration points but usually under half the gameplay points, and their games reach the ending in only 16% of runs, while references built by engineers with coding agents score near the maximum, which places this artifact-to-play gap in the agents. No two engines rank configurations alike (Kendall's tau 0.14-0.62), and the same game usually scores higher on Unity and Unreal Engine, which check more before play. We release Game Completion Boost, an agent skill for requirement-driven playtesting and repair that raises the execution-based score of three models by 14.1-24.3 points.

Abstract

AI Coding Agents

Explore similar work

From Code to Playable Systems: CEGame-Bench for Long-Horizon Game Creation Agents

GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?

VibeGame: Prompt-to-Game Development with AI-Native Engine and Self-Evolving Adversarial Agent Team