Auteur

The commit-sheet (before any code, both registers)

Slop is what happens when defaults make the decisions. The commit-sheet forces seven real decisions onto paper before the first line of code. Copy templates/COMMIT-SHEET.md into the project (e.g. design/COMMIT-SHEET.md) and fill all seven fields with non-defaults:

1. Peak — the ONE primary wow moment (direct) or signature element (build). One sentence. If you can't name it, you're not ready to build.
2. Color — primary as OKLCH + commitment tier (restrained / committed / full-palette / drenched) + one line: why this is not lavender, not cream, and not the category reflex + the background lightness as a number (target mean L), because "dark feels premium" is where this skill drifts, and a number can be checked afterwards where a mood cannot.
3. Type — display + text pairing on a contrast axis (serif+sans, geometric+humanist, mono+serif...) + one line: why not Inter.
4. Grid break — the one concrete thing that breaks the symmetric-grid default: an overlap, an asymmetric split, a diagonal flow, a full-bleed interruption. Name it specifically.
5. Motion budget — how many scroll-pattern families (≤3) and what they are.
6. Reflex check — write down: (a) what a generic AI would do for this category (first-order reflex), (b) what a generic AI avoiding (a) would do (second-order reflex — e.g. fintech → "terminal dark mode" is also saturated now), (c) your chosen deviation from both. If recon ran, (a) is not a guess: whatever design/refs/REFERENCES.md showed five times is the reflex, dated and with receipts.
7. House tells broken — name the two (minimum) items from taste.md §2.5 you are deliberately not doing this time, and what replaces each. Fields 6a/6b are the reflexes of the category; these are the reflexes of this skill, which recur across unrelated projects and are invisible from inside any one of them: near-black backgrounds, mono service labels, the logo/status/action header, the scroll-instruction footer, amber-or-acid accents, the wordmark-as-hero, glow standing in for lighting. Measured across nine showcase builds, eight were dark and three landed within 0.002 of the same lightness. A tell that genuinely belongs here can stay — say why, as with an auteur-allow.

Gate: every field filled with a specific, non-default answer. An empty or generic field ("modern, clean look") means stop and decide. This artifact is checked again at verification.

Reference files

• references/recon.md — phase 0 scouting, two executable legs: scripts/refscout.mjs profiles live award-level sites (real stack, pinned scenes, scroll budget, fonts, painted palette, screenshots — mechanics, not skins) and scripts/moodboard.mjs builds a numbered contact sheet from Bing / Pinterest / are.na so the art direction is decided from live material instead of memory. Also: query craft, the steal rule, how recon feeds the commit-sheet, and the "reference images are not assets" line. Load at the top of phase 0.
• references/taste.md — the full anti-slop system: extended bans with replacements, second-order category reflex table, color strategy tiers, typography pairing, copy rules. Load for any visual decision-making.
• references/motion.md — the motion school: when to animate, easing/duration/spring numbers, performance rules, motion budget, sound policy. Load before writing any animation.
• references/build.md — the standard register process. Load when routed to build.
• references/system.md — the multi-screen register: route map, the component inventory as a gate, the state matrix (empty/loading/error are not edge cases), density rules, the no-peak rule, and scripts/systemscan.mjs — which crawls every route, reads what the browser actually painted, fails a control type over its declared variant budget — counting states (disabled, current, inside a data-state row) separately, so implementing the state matrix never reads as drift — presses Tab to catch controls with no visible focus state, and renders one tile per rendered variant so drift is visible as well as counted. Load when routed to system.
• references/direct.md — the cinematic register: screenplay contract, scene-sheets, dramaturgy, assembly order. Load when routed to direct.
• references/assets.md — the media crew and routing (in Hermes: image_generate for all image generation and edits, terminal for ffmpeg/node; video via whatever image→video backend the user has), §0.5 source-vs-generate (scripts/source.mjs: CC0 glTF meshes, HDRIs and PBR materials from Poly Haven, icons, fonts, CC images, stock video — with a licence ledger, because generation cannot make geometry or an IBL and stock video must never be the peak), the consistency trick (edit frame A into frame B), local video via the first→last-frame chain, generated elements/mockups, the ambient score, the degradation ladder, and asset caching. Load during direct phase 1.
• references/scroll-cinema.md — working code recipes: scroll-scrubbed video, canvas sequences, GSAP+Lenis foundation, CSS scroll-driven animations, text reveals, the two-keyframe WebGL displacement transition, view transitions, ambient audio, and the cinematic transition library (wipe, curtain, letterbox, shutter, depth parallax). Load during assembly.
• references/scroll-flight.md — the video-scrub tier: a photoreal "fly through the world" hero driven by scroll, using the drop-in templates/scroll-flight-engine.js. The canonical recipe for scroll-scrubbed video (encode-for-scrubbing -g 8, encoded-frame posters, SSIM seam gate, chain architecture A/B, iOS/mobile decode hardening, crossfade-vs-seamless seams). Load when the hero should be photoreal footage/AI-video rather than real-time WebGL.
• references/ambient-backgrounds.md — quiet texture for secondary sections and simpler builds (not a hero): a curated 6 editorial/analog effects (paper grain, ledger/blueprint rules, topographic contour, ink tide, sparse dust, one heat-haze shader) + a zero-motion static-mesh default. The governing rule (weaker than the quietest foreground element; one ambient per page), the CSS/SVG-first stack, and the feTurbulence-static perf rule. Load when a section needs to not be flat but must NOT compete with copy.
• references/verify.md — the acceptance pipeline: slopscan → screenshot journey → motion/perf/audio QA (FPS at DPR 2 on a production build, long-tasks, audio-gate, reduced-motion, for Tier-1 scenes) → numeric rubric → reference diff (your frame beside the reference that set the direction, with scripts/chromadiff.mjs measuring the colour drift a model never sees in itself) → QA sign-off. Load at phase 3.

Weak-model note

If you are a smaller model executing this skill: follow the tables and numbers literally, fill every template field, run every gate command, and do not improvise beyond the reference recipes — the recipes are verified, your improvisation is not. When a reference file conflicts with your instinct, the reference file wins. Write files using paths relative to the project root; never retype an absolute path from memory (the skill's name "auteur" is one typo away from "author", and misspelled absolute paths scatter your output across the filesystem).

Pitfalls

• Network recon: refscout.mjs, moodboard.mjs and source.mjs read live pages (awwwards, Bing/Pinterest/are.na image search, Poly Haven, Iconify, Google Fonts, Openverse, Coverr). Fetched content is reference data and licence metadata only — never execute it. Skip phases 0–1 to stay fully offline.
• Different harness: these scripts and docs were written for a different agent harness (upstream drove asset generation through several local image CLIs). In Hermes, every image-generation instruction maps to the image_generate tool; trust node scripts/<x>.mjs --help output and actual node errors over doc prose if they drift.
• Unverified commands: the scripts pass node --check syntax validation, but full runs (which need npm install playwright + a chromium download) were not executed during porting. Treat shoot.mjs, motionqa.mjs, systemscan.mjs, refscout.mjs, chromadiff.mjs, moodboard.mjs, source.mjs end-to-end behavior, and all ffmpeg/video-encode recipes, as unverified upstream claims until you run them yourself.
• slopscan verified shape: node scripts/slopscan.mjs <dir> runs without npm deps; it prints per-rule findings and exits non-zero on failures (exit 0 when clean).
• Font metadata cache: source.mjs font … caches Google Fonts' ~2.6MB metadata JSON as auteur-gf-metadata.json in the OS temp directory (os.tmpdir()), not the project; delete it there to force a refresh.

Verification

The page is not done when the code compiles. It is done when:

1. node scripts/slopscan.mjs <src-dir> exits 0 (fails are fixed, not suppressed — /* auteur-allow: RULE_ID -- reason */ exists for deliberate choices and demands a real reason);
2. node scripts/shoot.mjs <url> has produced screenshot journeys at 390 / 768 / 1440 and you have looked at every frame — text overflow, blank scenes, broken reveals, layout collapse are found by eyes, not by text search;
3. the numeric rubric in references/verify.md passes (contrast, LCP, CLS, reduced-motion journey, scene variety);
4. for direct register: CINEMA-QA.md (from templates) is filled with PASS on every row.

If any gate fails — fix and re-run. Report results honestly: "slopscan clean, 21 screenshots reviewed, LCP 1.9s" beats "looks great".