Claude Sonnet 5.5 on Build a full game (code, art, audio) with AI
AClaude Sonnet 5.5 · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-14 · by @bobsiopener
- Estimated cost
- $0.19
- Wall time
- 1m
- Human step-ins
- 0
- Outcome
- Pass
Compare with other models on this task →
Summary
Exact prompt
Game jam, scaled down: in one pass build a complete small arcade game as a single file index.html (no external assets or libraries). Theme: "a capybara crossing a river". Must include: a title screen with Start, keyboard controls, procedurally drawn art on a <canvas> (no image files), at least 2 sound effects and background music generated with the Web Audio API (start audio on user gesture), a score, increasing difficulty, a game-over screen with restart, and a high score saved in localStorage.
Task prompt given to the agent (headless, offline, synthetic fixture inputs): Game jam, scaled down: in one pass build a complete small arcade game as a single file index.html (no external assets or libraries). Theme: "a capybara crossing a river". Must include: a title screen with Start, keyboard controls, procedurally drawn art on a <canvas> (no image files), at least 2 sound effects and background music generated with the Web Audio API (start audio on user gesture), a score, increasing difficulty, a game-over screen with restart, and a high score saved in localStorage. Inspired by https://x.com/qusaxz/status/2077427986689843623 (unverified; not checked with the author). This is an independent re-run on a reproducible synthetic task; it does not report or reproduce the original post's numbers. Cost shown is a list-price API-equivalent estimate (run on a Claude Pro subscription, not billed).
Success criteria: All automated rubric checks pass: single file, no external libs/assets; canvas drawing; Web Audio; oscillator music/sfx; title/start screen; game over + restart; localStorage high score; capybara theme; no uncaught console errors.
Setup notes: Cost is an ESTIMATE: Claude Code's list-price API-equivalent figure. The run used a Claude Pro subscription, so nothing was billed. input_tokens = all input incl. cache-creation and cache-read tokens; cached_tokens = cache reads (subset of input). Claude Code's cost bills cache writes at the 1h cache-write premium, so it can differ from a flat input-price recompute. Web disabled; tools Read/Write/Edit/Glob/Grep/Bash; single run; scored automatically (9/9 checks).
Estimated cost List-price estimate; not actually billed.
Metrics
- Estimated cost
- $0.19
- Wall time
- 1m
- Human step-ins
- 0
- Turns
- 3
| Model | Input tok | Output tok | Cost |
|---|---|---|---|
| AClaude Sonnet 5.5anthropic/claude-sonnet-5.5 | 75,671 | 11,377 | $0.19 |
Reproduce this run
- Use the use case prompt on the use case page.
- Same setup: Claude Sonnet 5.5 (anthropic/claude-sonnet-5.5) via
Claude Code 2.1.299 (headless -p), cloud. - It counts as a success only if it meets every success criterion: All automated rubric checks pass: single file, no external libs/assets; canvas drawing; Web Audio; oscillator music/sfx; title/start screen; game over + restart; localStorage high score; capybara theme; no uncaught console errors.
- Log model, tokens, wall time, human step-ins and the evidence log, then submit it as a reproduction. A different model? Submit a head-to-head rerun.
How runs are scored: Methodology.
Was this useful?
Automated checks
Trust 80% comes from these checks (how Trust % works).
- Run date is not in the future : passed
- Model is in the dated pricing table : passed
- Cost recomputed from tokens matches : passed
- Cached tokens are consistent : passed
- Output throughput is plausible : passed
- Autonomy matches step-ins : passed
- Step-ins do not exceed turns : passed
- Retries agree with first-try result : passed
- Evidence log SHA-256 matches : passed
- Evidence log is substantial : passed
Evidence log
{
"duration_api_ms": 75819,
"stop_reason": "end_turn",
"session_id": "[REDACTED]",
"total_cost_usd": 0.18528119999999998,
"usage": {
"input_tokens": 6,
"cache_creation_input_tokens": 16393,
"cache_read_input_tokens": 59272,
"output_tokens": 11377,
"output_tokens_details": {
"thinking_tokens": 1983
},
"server_tool_use": {
"web_search_requests": 0,
"web_fetch_requests": 0
},
"service_tier": "standard",
"cache_creation": {
"ephemeral_1h_input_tokens": 16393,
"ephemeral_5m_input_tokens": 0
},
"inference_geo": "global",
"iterations": [
{
"input_tokens": 2,
"output_tokens": 550,
"cache_read_input_tokens": 28739,
"cache_creation_input_tokens": 193,
"cache_creation": {
"ephemeral_5m_input_tokens": 0,
"ephemeral_1h_input_tokens": 193
},
"type": "message"
}
],
"speed": "standard",
"fallback_credit": null
},
"modelUsage": {
"claude-sonnet-5-5": {
"inputTokens": 6,
"outputTokens": 11377,
"cacheReadInputTokens": 59272,
"cacheCreationInputTokens": 16393,
"webSearchRequests": 0,
"costUSD": 0.18528119999999998,
"contextWindow": 1000000,
"maxOutputTokens": 128000,
"thinkingTokens": 1983,
"canonicalModel": "claude-sonnet-5-5",
"provider": "firstParty",
"costBasis": "list"
}
},
"permission_denials": [],
"terminal_reason": "completed",
"fast_mode_state": "off",
"fast_mode_disabled_reason": "sdk_opt_in_required",
"subagent_stats": {
"spawned": 0,
"requested": {
"background": 0,
"foreground": 0,
"unset": 0
},
"started_in_background": 0,
"max_depth": 0,
"spawned_by_subagents": 0,
"completed": 0,
"failed": 0,
"killed": {
"parent": 0,
"user": 0,
"system": 0
},
"refused": {
"depth_limit": 0,
"concurrency_limit": 0,
"budget": 0
},
"by_type": {}
},
"safety_stops": 0,
"is_error": false,
"num_turns": 3,
"subtype": "success",
"api_error_status": null,
"result": "I built the game as a single file, `./usecase-runner/runs/17-sonnet/work/index.html`. I only checked that the JavaScript parses. I haven't opened it in a browser, so gameplay, sound and the localStorage high score are untested.\n\n**How it plays**\n- **Goal:** the capybara hops from the bottom bank across 8 river lanes to the flowery bank at the top.\n- **Controls:** arrow keys or WASD to hop. P pauses and M mutes. Enter, Space or a click starts and restarts.\n- **Dying:** you lose a life if you land in water, ride off-screen or are on a turtle when it dives. You have 3 lives.\n- **Platforms:** logs drift in alternating directions. From level 2, some lanes have turtles that sink and resurface.\n- **Difficulty:** each level makes the lanes faster, shortens the logs and widens the gaps. The music tempo also speeds up.\n- **Score:** 10 \u00d7 level for each new row reached, plus 100 \u00d7 level for each crossing.\n- **Screens:** the title screen has a Start button. The game-over screen shows your score, the high score and a Restart button. The high score is saved in `localStorage`.\n\n**Art and audio**\n- **Art:** everything is drawn on the canvas, with no image files. That includes the capybara, logs, turtles, animated water, reeds and splash particles.\n- **Sound effects:** hop, splash, goal jingle, game-over jingle and a start tick, all synthesized with the Web Audio API.\n- **Music:** a looping pentatonic melody with a bass line.\n- **Audio start:** audio starts on the first key press or click.",
"ttft_ms": 21437,
"type": "result",
"duration_ms": 76297,
"uuid": "[REDACTED]",
"ttft_stream_ms": 1668,
"time_to_request_ms": 312,
"first_content_frame_ms": 1668,
"queued_turn_count": 0,
"result_index": 0
}