Taut Line

Build a tiny browser game called "Taut Line".

Game spec
---------
A deep-sea line-fishing cartridge built around a single simulated quantity: line tension. The line is a spring-damper along one axis that can only pull, never push, so a slack line exerts nothing and giving slack is a genuine defensive move rather than a slower reel. Reeling pays line in (raising extension and tension); slack pays it out (dropping tension but ceding distance). The fish is a point mass under thrust, water drag and line pull, fighting in bursts separated by rests, with shorter weaker runs as stamina falls. Spring, damping and the run-onset ramp are fixed by reaction-time measurement, not taste: they were tuned so the landing rate degrades smoothly from 100% at 120ms to roughly 40% at 450ms, because the first values let tension cross zero-to-break in 0.29s and lost 19 of 20 fights at an ordinary 260ms human reaction while every probe that re-decided the meter per frame reported a finished game. Three loss modes, each asserted by mechanism and not just outcome: the line parts at breakAt, sustained load above a fixed HOOK_LOAD for D.heat seconds tears the hook out, and a fish given more than its run-out allowance past the hook point escapes. Thrust is a fraction of breakAt scaled by a per-species guts factor, because an absolute thrust let the heaviest fish out-pull the line so no legal tension could hold it, and because without per-species guts the heavy fish were merely longer fights and always-casting-max was strictly dominant. Water drag and the resting brace are set so bulldozing loses every fight while reacting to the meter lands every species at every difficulty on every cast power — asserted by the tuning probes. Cast power (one shared integrator drives both the lure's flight and the dotted preview, so the preview cannot lie) sets both landing distance and which species bite, mapping reach to a biased slice of the roster so the top species is reachable but not free. Landing at higher tension and landing a fresh fish pay more; casts are limited per run. The HUD is honest: the tension meter reads the same value the break test uses, the hook-load limit is drawn as a notch (it is a rate limit the player cannot otherwise observe), and the run-out bar measures distance given since the hook rather than absolute distance, which previously showed a far cast as already lost before the fight began.

Hard constraints — the game will be embedded in a sandboxed iframe on a static gallery site, and a headless browser will screenshot it:

1. Output exactly ONE self-contained HTML file. All CSS and JS inline. Zero external requests: no CDNs, no external fonts, images, or sounds, no fetch/XHR/WebSocket. Any sound must be synthesized with WebAudio.
2. The iframe sandbox is "allow-scripts allow-pointer-lock" (plus an autoplay permission). So: no localStorage, cookies, or indexedDB (opaque origin — access throws); no popups, window.open, or navigation; no access to the parent page; alert()/prompt()/confirm() are silently blocked.
3. The playfield must be visible within 800 ms of load — that is when the screenshot is taken. No dead white screen. If there is a start screen, make it visually rich.
4. Controls: keyboard AND mouse must both work, plus on-screen touch buttons for phones. Call preventDefault() on game keys (arrows, space) so the page never scrolls.
5. The view is 16:9 and must scale to fill the viewport with no overflow or scrollbars.
6. Show a visible score. If the genre has a losing state, show a game-over screen with a restart (button and a key). Idle/clicker games skip game over but still need clear numeric feedback.
7. If you track a score, report it to the arcade shell so the player's best is remembered between visits: on every score change, call parent.postMessage({ source: "ai-arcade", type: "score", score: <number> }, "*"). It is safe to call it no matter what the parent does with it.
8. Keep it tiny: aim for under 400 lines, 60 fps, no frameworks. That 400-line target does not apply to 3D games — true 3D (a camera and perspective-projected world) or pseudo-3D (projected road/track, depth-sorted sprites) — which need their own software renderer and typically land at 800–1400 lines. The rest of this constraint still binds every game: 60 fps, no frameworks, no external assets, and no padding to reach a length.
9. Difficulty: the arcade shell passes the player's chosen difficulty as a URL query parameter. Read it with new URLSearchParams(location.search).get("difficulty") — the value is "easy", "medium", or "hard" (missing or unknown means "medium"). Make the three levels meaningfully different — speed, lives, enemy pressure, reaction time, whatever fits the genre; clicker/idle games with no losing state use starting resources and upgrade cost curves instead. Show the active level in the HUD. Without the parameter the game must be fully playable at its default (medium) tuning.

Return only the complete HTML file, no commentary.