Youssef
5b4bf7e443
Four things the last pass left open. Reflex findings no longer masquerade as Floor failures. They report at a new REFLEX grade which still blocks, so nothing drifts silently, but "0 FAIL" is now a true statement about the floor even while a build owes an argument above it. Across every committed page that distinction turns 230 reported failures into 56 real Floor findings and 174 arguable ones, which is the number that was hiding. The waiver cap is gone. Three-per-artifact was the restraint the tier existed to remove: a colour-field poster that wants a black stage, achromatic neutrals, a long measure and one face doing everything is right, and it was being blocked for being expressive. Guards, reason quality and the log are what keep waivers honest. Past three the build gets a NOTE saying what it now is, a house style rather than a set of exceptions, and pointing at design.md. The Floor has two admission criteria instead of one embarrassed exception. Door 1 is "no admired page breaks this on purpose". Door 2 is a named house rule where admired pages do break it but a language model reaches for it unprompted on every section. Gate 54 is the only member, listed in a table that any future addition has to join explicitly. The apology is gone; the reasoning is in the file. Finish findings gained a record form. `answered <check> · <reason>` parses, so a build that reasons its way past gate 38 leaves something the next run can read instead of prose in a comment. Also: the skill stops mandating em and en dashes in page copy, which had it arguing with its own house rule. The Picks block and the Step 5 preview merge into one recital, since the same decisions were being serialised four times. The font catalog and the per-tone voice samples become skip-unless-needed, roughly 200 lines a build. Two waiver bugs found by a cold build are fixed: "intentional design choice" was passing the boilerplate test, and a malformed two-field waiver was being ignored in silence. The 18 hero captures and 14 gallery thumbnails are regenerated against the live pages, so the README and the homepage stop advertising eyebrows the skill bans.
4.8 KiB
Roadmap
What's next. (v1.2 shipped: four new themes with the image hook, spec files for all 24 themes, the edit-time lint hook, the Tier A conformance harness, variants v2 with progressive reveal + grafts + thumbnails + scoped injection and the corner dock, and the harness pivot - one core with per-harness adapters for Claude Code / Codex / OpenCode plus the multi-harness installer.)
Since v1.2: the floor / library split. The eyebrow is banned outright, in every geometry, with a DOM-shape detector behind it and twelve replacement openings in section-entry.md. The 58 gates gained tiers: 34 carry a Floor clause that nothing ships through, 18 are Reflex gates a build can waive on the record with a mechanically re-derived guard and no cap, 6 never block. FAIL and REFLEX are separate grades, so "0 FAIL" is a true claim about the floor even while a build still owes an argument above it. Themes were cut to a written contract and are material rather than page plans, which took them from 4,070 lines to 2,861. The catalogs stopped being a mandatory pick and became a place to look.
Now
Pay down the floor debt in the committed corpus. Now that the gates are tiered, the real number is visible: across every committed page there are 56 Floor findings, not the 230 the old flat linter reported. They cluster as 18 token improvisations (gate 48), 9 layout animations (14), 8 pages missing overflow-x: clip (34), 4 missing reduced-motion fallbacks (27), 4 missing stamps (20), 4 purple gradients (2), 3 long-word wraps (51), 3 missing focus rings (26), 2 double-sticky (56) and one placeholder name (19). Several are access failures and none of them is taste. This is the highest-value cleanup left.
A Floor violation sits on the flagship example. site/examples/tally italicises one word inside its roman hero headline, which is gate 38a-i, and that page is the first card in the README table. Either fix the headline or stop leading with it.
Conformance on all three harnesses. Tier A (gen-cli.mjs) drives Claude Code today; run it in a logged-in terminal. Next: the same conformance pass through codex exec and OpenCode's non-interactive run, asserting each harness loads its adapter (harnesses/codex.md / harnesses/opencode.md), holds the load-order discipline, and lands 0-FAIL sloplint. (The old Tier B bare-API open-model matrix is retired from the roadmap; the tooling stays in eval/ as an archive.)
Measure the variants speedup. eval/variants-bench.md documents the four-config protocol; run it on a real authenticated machine and publish the wall-clock + token table (parallel + sketch + progressive vs the v1 baseline).
Second example builds for the four new themes. Each shipped with a canonical -01; add a -02 on a second brief, and marketing cards on the homepage gallery.
Next
sloplint as an edit-time hook, on by default. The hook exists (scripts/install-hook.mjs) and is opt-in. Consider shipping a project .claude/settings.json with it pre-wired, or a one-line prompt in init.
Grow the direction atlas + rate the draw. More lineages, usage-weighted dealing, a ratings pass so strong directions surface more often.
Variants v2.1. Screenshot thumbnails in the picker are in; next: DOM-injection with write-back for non-file-router apps (the hard AST-codemod path deliberately deferred), and a live variants demo wired on usehallmark.com.
Brand-first flow. From a short product description, generate a complete brand (palette, type, voice, imagery via the image hook) locked into a design.md; then every page builds against it.
Theme-aware motion tokens. Per-theme --dur-micro / --dur-short / --dur-long; Atelier should feel slower than Brutal.
Multi-page coherence. Lock the brand axes, vary the page-voice axes: different pages of the same site, not different sites.
study reads your own codebase. Third input mode: a path. Walk the files, extract the tokens + fingerprint in use, emit the same design.md.
Charts reference. A data-viz.md: small multiples over dense singles, one accent + neutrals, the worst chart types banned outright.
Later
- More harness adapters. Claude Code, Codex, and OpenCode are first-class today; Gemini CLI / Copilot adapters only if demand shows up, same one-core-plus-adapter shape.
hallmark explain- narrate the choices axis by axis; the skill teaches.- Negative-capability rules - for each anti-pattern, the perceptual reason it fails.
- Emotion-first prompting - nostalgic, optimistic, sceptical instead of tone adjectives.
- Sound + haptic policy - when web sound is acceptable without kitsch.
- Live preview as an MCP server - watch, render, screenshot, feed back into the gate sweep (the
--rendertier andgen-cli.mjsare the seeds of this).