Agents In UseInteractive agents, game experiments and simulations, including cases where the result is a playable prototype or a tested strategy.
66 use casesShowing 25–48
Preparing search…
Claude Code tuned opposing faction players through headless simulations. Human rule fixes and turn GIF inspection checked active play alongside near-even wins.
Added
Claude
kirsantov.comClaude Code built a seeded Godot loop prototype. Actual editor testing caught a built-in type collision; game-feel review and balancing work remained ahead.
AddedClaudevia Claude Code
tapestria.questClaude Code project agents built a standalone RPG evaluator with independent tests. Seeded scenarios exposed outdated rules; Foundry parity was not claimed.
AddedClaudevia Claude Code project; separate model-family review
rossgarlick.comClaude Code and Codex built a startup-ranking game. Review moved ELO to the server and fixed reset timing while the owner chose mechanics and launched a beta.
AddedClaudevia Claude Code; Codex review; Cowork planning
loopedin.gamesClaude Code built a Unity bullet hell using scene and play-mode tools. Human code and asset review supported an itch.io release; Steam remained future work.
AddedClaudevia Claude Code with OpusClaude Code simulated a hexcrawl's encounter curve and revised color behavior. Human play and player feedback changed the results despite passing tests.
AddedClaudevia Claude Code with Godot MCP
alsade.meClaude built three chess prototypes from one prompt; changing installed skills altered rules, layout, and architecture in an informal artifact comparison.
AddedClaudevia Claude Code
medium.comClaude Code and Codex played native macOS chess through accessibility labels and MCP. A deterministic referee rejected illegal moves before real UI actions.
AddedClaudevia Claude Code bridge; Codex opponent
koffeecup.netClaude Code helped port an Unreal game to iPhone. Device tests exposed video, build, input and material problems; TestFlight distribution remained unfinished.
AddedClaudevia Claude Code
nilssanderson.comClaude Code with Luna built a Godot rugby runner through small edits. Real engine errors and human play checks guided movement rather than broad game requests.
AddedClaudevia Claude Code with Luna via OpenRouter
note.comClaude rebuilt a Godot rotation-battle prototype after its creator found it had used chat notes instead of specs, exposing a polished, incorrect first screen.
AddedClaudevia Claude Code; Gemini supplied early design discussion
omar-alhaddad.comOmar Alhaddad shipped two mobile puzzle games with Claude, iterated on shaders and used custom skills for achievement icons and store screenshots he reviewed.
AddedClaudevia Claude Code
devpost.comCodex built a live robot mission simulation with checked routes, memory, and audit evidence; its owner kept optional physical camera access strictly read-only.
Added
Codexdevpost.comCodex built Neural Apex’s strategy-run experiment with telemetry and local explanation fallback; its source labels AI mechanics as educational approximations.
AddedCodex
developers.openai.comCodex built a space game linking flight, landing, walking and takeoff. Journey tests checked transitions while the maker judged controls and rendering limits.
AddedCodexvia Codex with Astradevpost.comCodex built a two-level WebMCP puzzle with unequal human and agent information, dynamic tools, and reconnection; its creator retained art and playtesting.
AddedCodexdevpost.comCodex built an offline causal-chain game with order-sensitive scoring and trap penalties; missing model access falls back, while live multiplayer stays future.
AddedCodexdevpost.comCodex helped improve Ghostline’s Android race and timing-sensitive checkpoint detection; Pattanaik checked actual APKs after stale builds confused verification.
AddedCodexvia Codex with GPT-5.6devpost.comCodex coordinated four simulated rescue robots through WebMCP, with one human-approved mission, temporary execution authority, a receipt, and tool revocation.
AddedCodexvia Codex in-app browser with WebMCPdevpost.comCodex helped build a continuous flight evaluation after paused-time cheating; three blind simulated landings retained full tool and timing evidence for review.
AddedCodexvia Codex and WebMCPdevpost.comCodex repaired Faultline’s replay and daily-reward edge cases, separating a dated ledger from bounded history and checking storage and keyboard behavior.
AddedCodexvia GPT-5.6 Sol; ChatGPT supported original conceptdevpost.comCodex built Cold Case with backward crime generation, unique-culprit checks, and one seeded random stream, publishing a replayable offline mystery game.
AddedCodexvia GPT-5.6devpost.comCodex hardened P.A.T.H. Vietnam’s scene continuity, concurrent saves and replay, while Robert Scales retained historical authorship and release review.
AddedCodexvia Codex GPT-5.6; OpenClaw supported earlier researchdevpost.comCodex helped separate race and scenery randomness for reliable personal-best ghosts, then checked mobile behavior and local narration in a static browser game.
AddedCodex