wiki: add Getting Started, FAQ, Anti-Patterns Reference; refresh Home to v3.1.0
- Home: update to v3.1.0, document all 12 manifests and the 18-tool MCP server.
- Getting-Started: install on 17 IDEs, the discover/recommend/design/lint loop, CI gate.
- FAQ: pre-install questions, how it differs from a prompt or token kit, what runs local.
- Anti-Patterns-Reference: the 152 rules by category and severity, with reasoning and fixes.
All counts verified against `ux stats` at v3.1.0. No em-dashes, no emoji.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
4535d94
v3.0.0 wiki — Release Notes v3 (with T's bug fixes documented)
27f05c0
v3.0.0 wiki — add Three-Modes-Explained + Migration-v2-to-v3
9b59d58
v3.0.0 wiki — rewrite Home page for v3 + add Axis Interaction Matrix
9fdbc68
v3.0.0: add 4 wiki pages — The Brain, Synthesizer, Decisions Ledger, Evolve Loop
6ed1644
v2 wiki: 6 new pages — Architecture, Recommender-Engine, Linter-Rules, Brand-Library-110, Anti-Pattern-Catalog-100, Cross-IDE-Distribution
515e7e1
v2 wiki refresh — 4 core pages updated for the current state
The wiki was on v1 (cream/coral, no Python engine, 17 commands, 60+
pattern arsenal, 30+ reference files). Refreshed the 4 most-visited
pages to v2 reality:
- Home.md — wiki landing. Now lists current numbers (100 rules, 110
brands, 22 commands, 17 IDEs, 75 tests, 17-language website, MCP
with 14 tools). Live competitor comparison table. Star history.
- Installation.md — three install paths (pip / npx / Claude Code).
17-IDE adapter table with target flags. MCP install + verify.
- MCP-Server.md — new page. 14 tools listed, config snippets for
Claude Desktop / Cursor / Windsurf / generic hosts. Example tool
call with response. Competitor comparison (none ship MCP).
- All-22-Commands.md — new page (replaces All-17-Commands.md). Every
command organized by workflow stage. Chaining example.
0a0b71b
feat: ship 25-page SEO-targeted wiki for ux-skill plugin
Tier 1 — Core reference (5): Home, Installation, All-17-Commands,
All-5-Sub-Agents, Discovery-Protocol — 1,516 lines
Tier 2 — Builder how-tos (5): fix AI-generated UI, build SaaS landing,
generate design system, audit accessibility, fix microcopy — 2,835 lines
Tier 3 — Advanced how-tos (5): design dashboard, motion + CWV, detect
AI-slop, AI to human-grade, case study from product data — 2,826 lines
Tier 4 — Authority canon (5): 30 Laws of UX, Norman's principles, Krug's
3 laws, anti-AI-slop ban list, creative arsenal — 2,408 lines
Tier 5 — Process + comparison (5): 6 audit lenses, Polaris reports,
Wfrah case-study format, frontend stacks compared, real-life UX
consulting — 3,405 lines
Total: 12,990 lines across 25 pages. SEO-targeted at high-search-volume
long-tail keywords across builder pain points + authority canon +
buyer-stage comparisons. Every page: snippet-friendly opener, internal
wiki cross-links, footer with repo + LinkedIn for backlinks.
e9d19be