No artificial mount-time loading delay — drive skeletons from real async. useEffect(() => { setTimeout(() => setLoading(false) AI frontend skill pack Registry-routed. Machine-enforced. Anti-slop by default — a request loads exactly one of 19 skills, and everything it writes is held to 60 checks before it ships. Scroll to explore 01 — The problem Behind the router sits 372,212 tokens of reference depth — material every skill can point into, none of it loaded until a request routes there. The pack is not handed to an agent whole: a router matches your request against trigger keywords, opens exactly one of 19 skills plus the core files it declares, and everything else stays on disk. 02 — How it works Matched by keyword, not guessed The router reads your prompt, matches it against every skill's trigger keywords, and loads exactly one — the most specific match — plus the core files it declares. No match, and it asks a clarifying question instead of guessing. Try it below. Checked before it ships What gets written is held to 60 machine-checked constraints (17 AST + 43 regex) — the first half walked through the TypeScript compiler API, the second run as patterns. Ten deliberate anti-examples exist to prove the checks actually fire. Only what's needed loads One skill plus its declared dependencies — nothing else opens. A typical request loads a few thousand tokens against the full depth available, and every shipped example compiles under tsc --strict. Heroes, pricing, testimonials, bento grids, logo walls, comparison tables, FAQ, CTAs, footers — plus empty states and onboarding narrowing the cost372,212 5,999 tokens — 1.6% of the depth available this request 5,999 of 372,212 03 — The wall Walked through the TypeScript compiler API rather than matched as text — the eight rules a regex cannot see without understanding the tree it's reading. Run as regular expressions over source. Faster to check, and the half this page's own checker below can run in your browser. No...