LAWSOFUX · motion lab
Colophon · build record

How this lab notebook was built.

A short provenance note. The site is one output of a three-phase autonomous run: ontology authoring, Cavalry recreation, and lab-notebook publishing. Every artifact is reachable from the FEATURE_FLAGS manifest at the project root. Reverting the integration takes four commands.

Source

The site files live under the ShurIQ design system project, isolated to a single folder.

projects/shur/shuriq-design-system/ontology-laws-of-ux-v1/
├── cards/                    30 concept cards (markdown + frontmatter)
├── svg-fallbacks/            30 inline animated SVGs
├── cavalry-scenes/           30 Cavalry scene scripts
├── research/                 video cadence notes, frame map
├── graphs/                   InfraNodus graph exports
├── site/                     this site (build.py + handwritten chrome)
├── FEATURE_FLAGS.md          isolation manifest
└── README.md                 project overview

Build pipeline

The thirty concept pages and the gallery index are emitted by site/build.py — a Python 3 standard-library script with no external dependencies. The script reads each concept's card markdown, animated SVG, Cavalry scene, and verified cadence notes, then templates a self-contained HTML page. The four chrome pages — blog, rubric, primitive, about — are hand-authored. CSS lives in a single file at site/css/site.css.

Build is reproducible:

cd site
python3 build.py
# → built 30 concept pages + index.html

Agent team

The work was completed across three phases by autonomous agents, each scoped narrowly.

Phase 1 · Ontology authoring

Wrote thirty concept cards as structured markdown with frontmatter — definition, ShurIQ-relevance prose, takeaways, motion language, origins, related-concept wikilinks. Set up the FEATURE_FLAGS isolation manifest. Drafted two new agent skills: laws-of-ux for per-element diagnosis and design-rubric-complex-apps for whole-screen scoring.

Phase 2 · Cavalry recreation + verification

Audited a 24-minute screencast of the lawsofux.com site frame-by-frame at 0.5 fps. Produced verified motion cadence notes for all 30 concepts. For six concepts where video coverage was incomplete, fetched the inline hero SVG from the production page and rewrote the cadence prose with corrected geometry. Authored thirty Cavalry scene scripts using the validated shuriq-motion API surface. Built thirty inline animated SVG fallbacks. Surfaced the opacity-stepped-bands candidate primitive after noticing three concepts converged on the same scaffold.

Phase 3 · This site (you are here)

Read the Phase 1 + 2 outputs as inputs. Built the static site generator. Wrote the editorial blog post about the Cavalry MCP setup and the skills layer. Rendered the rubric as a fillable scoring template. Wrote this colophon. Deployed to Cloudflare Pages.

Isolation guarantees

The integration ships behind a single FEATURE_FLAGS manifest. Four guarantees hold:

If the experiment fails, removal is procedural:

rm -rf ~/.dotfiles/ai/skills/laws-of-ux
rm -rf ~/.dotfiles/ai/skills/design-rubric-complex-apps
rm -rf projects/shur/shuriq-design-system/ontology-laws-of-ux-v1
# delete InfraNodus graph: shuriq-design-ontology-v1
git revert <integration-sha>

Voice and design

The site is intentionally framed as lab notes rather than a product surface. The biological/regulatory analogy carries through: the ontology is a tissue sample of the design discipline; the rubric is a diagnostic instrument; the FEATURE_FLAGS manifest is a calibration record; the git revert is the explant procedure. The chrome stays still — typography, spacing, hairline rules — and lets the SVGs animate. Type stack is IBM Plex (serif for display, sans for body, mono for code). The cream palette (#FAF8F5) matches the SVG fills so animations sit on the page without a frame.

What this is not

This is not a derivative of lawsofux.com. The cards are paraphrased into ShurIQ-relevance prose, the SVGs are original recreations based on documented motion cadence, and the takeaways are tuned to the brand-intelligence context. The Cavalry scenes are written from scratch against the validated shuriq-motion API. Source attribution to lawsofux.com lives in each concept page's sidebar.

Contact

This is a Sense Collective lab notebook. The ShurAI team built it. Written feedback welcome through the channels the user already uses.