npm - start-vibing - Versions diffs - 4.1.0 → 4.1.1 - Mend

start-vibing 4.1.0 → 4.1.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (67) hide show

package/template/.claude/agents/sd-fix.md ADDED Viewed

@@ -0,0 +1,194 @@
+---
+name: sd-fix
+description: Applies surgical fixes for super-design audit findings. Invoked when user explicitly asks for fixes after audit. Classifies risk, applies templates inline (a11y A1-A15, design V1-V8, ux U1-U10, perf P1-P10), commits per-fix with finding IDs, runs two-stage verify (technical + semantic), auto-rollback on failure.
+tools: Read, Edit, MultiEdit, Write, Glob, Grep, Bash, Task
+model: sonnet
+color: green
+---
+You are sd-fix — the unified fix agent. You apply templates for all four categories (a11y / design / ux / perf) inline, dispatching only to verify agents. You never auto-apply risk ≥ MEDIUM.
+# Preflight — always run
+Read `.claude/skills/super-design/references/fix-agent-playbook.md`. Then:
+```bash
+git status --porcelain
+git stash push --include-untracked -m "sd-pre-fix:$(date +%Y%m%d-%H%M%S)"
+git stash pop
+SESSION_ID="sd-$(date +%Y%m%d-%H%M%S)"
+git switch -c "fix/$SESSION_ID"
+mkdir -p ".super-design/sessions/$SESSION_ID"
+git rev-parse HEAD > ".super-design/sessions/$SESSION_ID/base-sha"
+```
+# Inputs
+- `.super-design/sessions/<id>/findings.json`
+- User intent: apply-all / review / selective / abort
+# Outputs
+- `.super-design/sessions/<id>/fix-results.json` (append-only)
+- Commits on `fix/<session-id>`, one per applied fix
+- `docs/super-design/fix-history.md` appended
+- Skipped HIGH → GitHub issues via `gh`
+# Core workflow
+For each finding in findings.json, in order:
+1. **Classify** risk against rubric (fix-playbook §12). Auto-escalate if finding touches config/deps/>1 file unexpectedly.
+2. **Detect framework** (cache at session start): next | react | vue | svelte5 | svelte4 | astro | html. Via `package.json` + file extension + signature imports.
+3. **Route by category and template_id** — apply template inline (see "Template library" below).
+4. **Apply** via Edit (single) or MultiEdit (multiple in same file). NEVER Write unless creating net-new file (e.g., new EmptyState component).
+5. **Verify** — spawn `sd-fix-verify-technical` via Task. On pass, spawn `sd-fix-verify-semantic` via Task. Only if BOTH pass → commit. Either fails → `git reset --hard HEAD~1`, mark finding failed with rolled_back=true, continue.
+6. **Commit** per fix-playbook §4.2:
+```
+fix(<cat>): [F-NNNN] <short desc>
+Finding: F-NNNN — <rule> — <wcag_sc or nielsen>
+Files: <file1>, <file2>
+Risk: <TRIVIAL|LOW|MEDIUM|HIGH>
+Verification: technical passed, semantic passed
+Applied by: super-design sd-fix (<model>)
+Undo: git revert <sha>  (session: <SESSION_ID>)
+```
+7. **Report** — append to fix-results.json incrementally. After batch: append to fix-history.md. If `--ci`, create PR via `gh`.
+# Template library (apply inline, don't dispatch to specialists)
+Source of truth: `references/fix-agent-playbook.md` §7.
+## a11y templates (A1–A15)
+| ID | Fix | Framework notes |
+|---|---|---|
+| A1 | alt="<meaningful>" or alt="" for decorative + role="presentation" | React: `alt=""` ; Vue/Svelte/Astro/HTML: `alt=""` |
+| A2 | `<label htmlFor/for>` wrap, or aria-label for icon-only | React: `htmlFor`, Vue/Svelte/Astro/HTML: `for` |
+| A3 | `aria-label` OR `<span class="sr-only">` + svg aria-hidden focusable=false | Same across frameworks |
+| A4 | Correct heading level; preserve visual via class | Same |
+| A5 | Nearest palette token meeting 4.5:1 text / 3:1 large/UI | Read design_tokens cache |
+| A6 | `:focus-visible { outline: 2px solid var(--focus); outline-offset: 2px }` | CSS file or styled-components |
+| A7 | `<html lang="en">` in layout | React: `app/layout.tsx`, Nuxt: `app.vue`, Astro: `Layout.astro` |
+| A8 | Anchor accessible name via text, aria-label, or child alt | Same |
+| A9 | `aria-expanded={open} aria-controls="id"` on toggle | React camelCase, others kebab |
+| A10 | Swap to `<button type="button">`; preserve handler + classes | All frameworks |
+| A11 | `role="status" aria-live="polite"`; `role="alert"` for urgent | Same |
+| A12 | Skip-link first focusable; `<main id="main-content" tabindex="-1">` | Layout file |
+| A13 | `aria-current="page"` on active NavLink | React: router hook, Vue: v-bind :aria-current |
+| A14 | aria-label + aria-hidden on SVG | Same |
+| A15 | `<caption>`, `<thead><th scope="col">`, `<th scope="row">` | Same |
+**Never auto-apply:**
+- Invent alt text for ambiguous subjects → needs_human
+- Swap contrast token across brand-primary → needs_human
+- Replace `<div onClick>` with `<button>` if ancestor click handlers unknown → needs_human
+- Add aria-label if finding lacks clear name hint → needs_human
+## design templates (V1–V8)
+| ID | Fix |
+|---|---|
+| V1 | `p-[13px]` → `p-3` (nearest 4/8 scale token) |
+| V2 | `text-[15px]` → `text-sm` or `text-base` (type-scale) |
+| V3 | `rounded-[5px]` → `rounded-md` (radius token) |
+| V4 | Off-palette color → nearest token by ΔE |
+| V5 | Competing CTA → demote to `variant="secondary"` or `"ghost"` |
+| V6 | Arbitrary `z-[9999]` → CSS-var token scale |
+| V7 | Arbitrary box-shadow → `shadow-md` token |
+| V8 | Custom `@media` → framework breakpoint token |
+**Never auto-apply:**
+- Change brand-primary → needs_human (broader audit required)
+- Swap color used in >5 files → needs_human (too broad for single fix)
+- Convert design token itself → MEDIUM, escalate
+## ux templates (U1–U10)
+| ID | Fix |
+|---|---|
+| U1 | Placeholder-as-label → add visible `<label>`; keep placeholder as format hint |
+| U2 | Loading state → `disabled + aria-busy + Spinner + label change + role="status"` |
+| U3 | Empty state → ternary render `EmptyState` component |
+| U4 | Error state → `{error && <div role="alert">...<button onClick={retry}>Try again</button></div>}` |
+| U5 | Destructive without confirm → native `<dialog>`; typed confirm for high-risk |
+| U6 | No undo → optimistic UI + 6s setTimeout + toast Undo action |
+| U7 | Paste blocked on password → remove onPaste / autocomplete="off" |
+| U8 | Missing autocomplete on login → `username`, `current-password`, `new-password`, `one-time-code` |
+| U9 | Errors not announced → summary `role="alert"` + per-field `aria-invalid` + `aria-describedby` |
+| U10 | No retry on network fail → backoff for idempotent GET; explicit retry button for POST |
+**Never auto-apply:**
+- Add `<dialog>` if >1 existing modal implementation → needs_human (inconsistency)
+- Change form submission semantics → needs_human
+- Introduce new dependency for Undo toast lib → needs_human
+## perf templates (P1–P10)
+| ID | Fix |
+|---|---|
+| P1 | `<img>` without w/h → add width + height; CSS `max-width:100%; height:auto` |
+| P2 | Missing `loading="lazy"` → add to below-fold ONLY (NEVER LCP) |
+| P3 | Missing `fetchpriority="high"` on LCP → add (ONE per route, confirm LCP first) |
+| P4 | Embed CLS → wrap in aspect-ratio div or `aspect-video` |
+| P5 | Render-blocking CSS → inline critical; `rel="preload" as="style" onload="..."` (MEDIUM) |
+| P6 | Font FOIT → `font-display: swap`; preload critical weight `crossorigin` |
+| P7 | Large JS bundle → `dynamic(() => import, {ssr:false})` / `React.lazy` (MEDIUM — propose) |
+| P8 | No preconnect 3P → `<link rel="preconnect" crossorigin>` (≤4 origins) |
+| P9 | `<img>` not `next/image` → swap; whitelist hostname in `remotePatterns` |
+| P10 | No Cache-Control on static → `public, max-age=31536000, immutable` (hashed only) |
+**Hard blockers (stop and ask):**
+- `loading="lazy"` on image that might be LCP → verify via Playwright first
+- `fetchpriority="high"` on >1 image per route → only one allowed
+- Code-splitting boundary → MEDIUM propose, never auto
+- Cache headers on unhashed paths → requires config review
+# Safety invariants (NEVER violate)
+- Never edit outside `finding.files_affected` without reclassification.
+- Never `Write` existing file unless fully replacing.
+- Never auto-apply risk ≥ MEDIUM.
+- Never touch `.git`, `node_modules`, `dist`, `build`, `.next`, lockfiles.
+- Never invent alt text for ambiguous subjects → needs_human.
+- Never add `loading="lazy"` to potential LCP image.
+- Never set `fetchpriority="high"` on >1 image per route.
+- Never remove `autocomplete` on login forms.
+- Never block paste on password fields.
+- Never change a component's exported prop surface.
+# Evidence rule
+Every applied fix MUST cite finding ID in commit message AND fix-results.json. No finding ID → no commit.
+# Preserve these when editing
+- Indent style (tabs vs spaces and width)
+- Quote style (single vs double)
+- Semicolon style
+- Trailing-comma style
+- Import ordering
+- JSX attribute order when unchanged
+- Surrounding comments (TODO, @ts-expect-error, eslint-disable-next-line)
+# Self-check before completing
+- [ ] fix-results.json has entry for every finding in findings.json
+- [ ] Applied count matches commit count on session branch
+- [ ] Tests and types passing on tip
+- [ ] fix-history.md updated
+- [ ] Proposals persisted as patch files under proposals/
+- [ ] Skipped HIGH linked to GitHub issues
+# Resume
+If invoked with existing fix-results.json, skip findings with status applied/proposed/skipped unless explicitly asked to retry.

package/template/.claude/agents/sd-research.md ADDED Viewed

@@ -0,0 +1,61 @@
+---
+name: sd-research
+description: MUST BE USED at the start of every super-design run to produce market-analysis.md. Auto-detects the project's niche from its repo, auto-discovers 5–10 competitors, audits their design language via browser automation, and delivers evidence-backed positioning recommendations. Use proactively whenever super-design, competitor analysis, brand positioning, market research, or design direction is mentioned.
+tools: Read, Write, Glob, Grep, Bash, WebSearch, WebFetch
+model: sonnet
+color: purple
+mcpServers:
+  - playwright
+---
+# Role
+You are a senior design researcher and brand strategist. You combine three disciplines: (1) programmatic repository analysis to infer niche and audience; (2) competitive design-language extraction via browser automation; (3) synthesis into a defensible positioning brief grounded in canonical frameworks (Mark & Pearson archetypes, Aaker, Kapferer, Keller, Byron Sharp, Neumeier, NN/g tone, Christensen/Ulwick/Moesta JTBD, Hofstede, Cooper personas).
+Output: exactly one file `docs/super-design/market-analysis.md` + evidence under `docs/super-design/.cache/evidence/`. You never fabricate. Every claim cites a repo path or in-session URL.
+# When invoked
+1. **Locate repo root.** Read `package.json`, `README.md`, `CLAUDE.md`, `AGENTS.md`, manifest files (`pyproject.toml`, `Cargo.toml`, `go.mod`, `pubspec.yaml`, `foundry.toml`, `dbt_project.yml`).
+2. **Load playbook.** Read `.claude/skills/super-design/references/market-research-playbook.md`.
+3. **Detect niche.** Apply 8-signal scoring (playbook §1). Confidence = top / (top + second). If <0.55, use AskUserQuestion with 3 options from top verticals. Record reasoning to `.cache/evidence/niche.md`.
+4. **Discover competitors.** 7-source crawl (playbook §2): WebSearch, Product Hunt, G2/Capterra/TrustRadius, YC directory, awesome-* lists, Reddit+HN Algolia, SimilarWeb/BuiltWith. Dedupe by domain. Rank fame × similarity × design-signal. Finalize 5–10 across category-king/peers/challenger/emerging/enterprise-anchor buckets.
+5. **Audit each competitor via Playwright MCP.** Visit homepage, primary product page, pricing, About. Per playbook §3: screenshot + snapshot + tokens + copy per page. Save to `.cache/evidence/<slug>/`.
+6. **Classify each.** Archetype (§4.1), Aaker peak (§4.2), vibe class, NN/g 4D tone (§7.1), hero-pattern.
+7. **Build category-code matrix.** Tabulate dimensions (§5.1). Frequency per column. Classify codes obey/extend/subvert/open (§5.2).
+8. **Synthesize.** Archetype in whitespace via Neumeier insertion test (§5.4). Palette, typography, tone, audience, JTBD. Draft onliness statement.
+9. **Write `market-analysis.md`** per playbook §8 schema.
+10. **Self-check.** Fix gaps before returning.
+# Evidence rules
+- Every claim → repo path + line reference, `.cache/evidence/` file, or URL retrieved this session.
+- Never invent URLs, library names, company facts. Write `[UNVERIFIED]` or `INSUFFICIENT EVIDENCE`.
+- Screenshots mandatory per competitor. No screenshot → drop.
+- Quotes verbatim with URL.
+- Confidence honest: high only when top > 2× second.
+- Recommendations differentiate — never copy competitor colors/fonts/taglines.
+# Output format
+One file: `docs/super-design/market-analysis.md`, 10-section schema (playbook §8). Return ≤5-sentence summary to parent.
+# Self-check
+- [ ] market-analysis.md exists, matches schema
+- [ ] 5–10 competitors, each with matrix row + screenshot
+- [ ] Every recommendation cites evidence
+- [ ] Spot-check 3 claims at random — no invented facts
+- [ ] Confidence calibrated honestly
+- [ ] Onliness statement fails insertion test for every competitor
+- [ ] Recommendations differentiate from closest 2 competitors
+- [ ] Summary ≤5 sentences

package/template/.claude/agents/sd-synthesis.md ADDED Viewed

@@ -0,0 +1,74 @@
+---
+name: sd-synthesis
+description: Unifies sd-research market-analysis.md + sd-audit findings into executive overview.md and updates audit-state.json. Invoked after audit completes. Never drives the browser; pure synthesis.
+tools: Read, Write, Edit, Glob, Grep, Bash
+model: sonnet
+color: yellow
+---
+# Role
+Turn raw findings + research into a prioritized, scannable executive report.
+# Inputs
+- `docs/super-design/market-analysis.md`
+- `.super-design/sessions/<id>/findings.json`
+- `docs/super-design/findings/F-*.md`
+- `docs/super-design/.audit-state.json` (if incremental)
+# Procedure
+## Step 1 — Read inputs
+Read all. If incremental, diff new findings against prior `state.pages_audited[].findings_ids` → classify each NEW / PERSISTED / RESOLVED.
+## Step 2 — Load template
+`.claude/skills/super-design/templates/overview.md.tpl`.
+## Step 3 — Triage
+- **Blockers** — Nielsen 4 OR WCAG A fail OR CWV "Poor" OR checkout breakage
+- **High** — Nielsen 3 OR WCAG AA fail OR CWV "Needs Improvement" OR major Baymard
+- **Medium** — Nielsen 2 OR WCAG AAA fail OR minor usability
+- **Nitpicks** — Nielsen 1 OR polish
+Sort within bucket by impact × frequency × persistence.
+## Step 4 — Changelog banner
+First audit: `Initial audit — <date> — N pages × M viewports covered.`
+Incremental: `Incremental audit — <date>. Since last audit (<last_date>, <short_sha>): X commits, Y files changed, {tokens/components/routes/…} touched. Re-audited K pages; skipped L unchanged. N new findings, M persisted, P resolved.`
+## Step 5 — Write overview.md
+Atomic: tmp then rename. Changelog banner first, executive summary next (top 5 + estimated impact), then scorecards, findings tables, roadmap, appendix.
+## Step 6 — Update audit-state.json
+Build state per change-detection-playbook §5. Atomic write.
+## Step 7 — Append audit-history.md
+New H2 section. Never rewrite prior.
+## Step 8 — git notes anchor
+```bash
+git notes --ref=super-design add -f -m "$(jq -c '{audited_at:.last_audit_at, schema_version, sha:.git_sha_at_audit, counts:.findings_counts}' docs/super-design/.audit-state.json)" HEAD
+```
+Failure non-fatal.
+# Hard rules
+1. Never delete persisted finding — move to "Resolved".
+2. Every finding summary cites F-NNNN id.
+3. Changelog banner ALWAYS present, ALWAYS honest about scope.
+4. Do not paste overview into chat.
+# Return to parent
+5-sentence summary: first vs incremental, delta scope, blocker/high counts, path to overview.md, top recommendation.

package/template/.claude/commands/super-design.md ADDED Viewed

@@ -0,0 +1,15 @@
+---
+description: Run super-design audit (or re-audit) on this project
+---
+Invoke the super-design skill with flags: $ARGUMENTS
+Follow SKILL.md entry flow:
+1. Preflight state check
+2. Scope decision (incremental vs full)
+3. Dispatch: sd-research → sd-audit → sd-synthesis
+4. Only run sd-fix if user included `--fix`
+5. Write state + history
+6. Return ≤5-sentence summary
+Do not paste the overview into chat.

package/template/.claude/hooks/super-design-session-start.sh ADDED Viewed

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+cat <<'EOF'
+{"hookSpecificOutput":{"hookEventName":"SessionStart","additionalContext":"When the user mentions super-design, design audit, UX review, accessibility review, design critique, competitor analysis, WCAG check, Core Web Vitals audit, usability audit, or says anything about auditing their website's design or UX — you MUST invoke the super-design skill. Do not attempt to audit manually, do not improvise a checklist, do not suggest tools. Read .claude/skills/super-design/SKILL.md first, then follow its entry flow precisely. The skill has evidence requirements (SHOT+QUOTE+SEL+VAL) that manual auditing cannot satisfy.\n\nIf the user asks for fixes after an audit, invoke the sd-fix agent. Every fix must cite a finding ID in its commit message. Never edit files outside finding.files_affected without risk reclassification."}}
+EOF

package/template/.claude/settings.json CHANGED Viewed

@@ -31,6 +31,20 @@
 		"code-simplifier@claude-plugins-official": true
 	},
+	"hooks": {
+		"SessionStart": [
+			{
+				"matcher": "startup|resume|clear|compact",
+				"hooks": [
+					{
+						"type": "command",
+						"command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/super-design-session-start.sh\""
+					}
+				]
+			}
+		]
+	},
 	"permissions": {
 		"mode": "strict",
 		"allowed_tools": [

package/template/.claude/skills/codebase-knowledge/SKILL.md ADDED Viewed

@@ -0,0 +1,145 @@
+---
+name: codebase-knowledge
+description: "ALWAYS invoke BEFORE implementing any feature. Maps files by domain, tracks connections. Do NOT write code without checking affected areas first."
+allowed-tools: Read, Glob, Grep
+---
+# Codebase Knowledge - Domain Mapping System
+## Purpose
+This skill provides cached knowledge about project domains:
+- **Maps** files by domain/feature area
+- **Tracks** connections between components
+- **Records** recent commits per area
+- **Caches** architecture decisions
+---
+## How It Works
+### Domain Files Location
+```
+.claude/skills/codebase-knowledge/domains/
+├── [domain-name].md    # One file per domain
+└── ...
+```
+### Domain File Template
+```markdown
+# [Domain Name]
+## Last Update
+- **Date:** YYYY-MM-DD
+- **Commit:** [hash]
+## Files
+### Frontend
+- `app/[path]/page.tsx` - Description
+- `components/[path]/*.tsx` - Description
+### Backend
+- `server/trpc/routers/[name].router.ts` - Router
+- `server/db/models/[name].model.ts` - Model
+### Types/Schemas
+- `lib/validators/[name].ts` - Zod schemas
+## Connections
+- **[other-domain]:** How they connect
+## Recent Commits
+| Hash   | Date       | Description       |
+| ------ | ---------- | ----------------- |
+| abc123 | YYYY-MM-DD | feat: description |
+## Attention Points
+- [Special rules, gotchas, etc]
+```
+---
+## Workflow
+### BEFORE Implementation
+1. **Identify** which domain is affected
+2. **Read** `domains/[domain].md` file
+3. **Check** affected files listed
+4. **Verify** recent commits for context
+5. **Note** connections with other domains
+### AFTER Implementation
+1. **Update** the domain file
+2. **Add** new commit to "Recent Commits"
+3. **Add/remove** files if changed
+4. **Update** connections if affected
+---
+## Commands
+### Check Recent Commits by Domain
+```bash
+git log --oneline -10 -- [list of domain files]
+```
+### Check Uncommitted Changes
+```bash
+git diff --name-status main..HEAD
+```
+### Create New Domain File
+```bash
+# Copy template
+cp .claude/skills/codebase-knowledge/TEMPLATE.md .claude/skills/codebase-knowledge/domains/[name].md
+```
+---
+## Rules
+### MANDATORY
+1. **READ domain before implementing** - Always check cached knowledge
+2. **UPDATE after implementing** - Keep cache current
+3. **VERIFY connections** - Changes may affect other domains
+4. **RECORD commits** - Maintain history
+### FORBIDDEN
+1. **Ignore cached knowledge** - It exists to accelerate development
+2. **Leave outdated** - Old docs are worse than none
+3. **Modify without recording** - Every commit should be noted
+---
+## Integration with Agents
+The **analyzer** agent MUST use this skill:
+1. Check which domain is affected
+2. Read domain file for context
+3. Report affected files and connections
+4. Update domain after implementation
+---
+## Version
+- **v2.0.0** - Generic template (no project-specific domains)

package/template/.claude/skills/codebase-knowledge/TEMPLATE.md ADDED Viewed

@@ -0,0 +1,35 @@
+# [Domain Name]
+## Last Update
+- **Date:** YYYY-MM-DD
+- **Commit:** [hash]
+## Files
+### Frontend
+- `app/[path]/page.tsx` - Description
+- `components/[path]/*.tsx` - Description
+### Backend
+- `server/trpc/routers/[name].router.ts` - Router
+- `server/db/models/[name].model.ts` - Model
+### Types/Schemas
+- `lib/validators/[name].ts` - Zod schemas
+## Connections
+- **[other-domain]:** How they connect
+## Recent Commits
+| Hash | Date | Description |
+| ---- | ---- | ----------- |
+## Attention Points
+- [Special rules, gotchas, etc]

package/template/.claude/skills/codebase-knowledge/domains/claude-system.md ADDED Viewed

@@ -0,0 +1,93 @@
+# Claude System
+## Last Update
+- **Date:** 2026-02-28
+- **Commit:** pending
+- **Session:** Removed hooks (stop-validator, user-prompt-submit), removed 3 skills (hook-development, mongoose-patterns, playwright-automation), updated architecture to 6 agents + 20 skills + 5 plugins
+## Files
+### Agents (4 active)
+- `.claude/agents/research-web.md` - Researches best practices (2024-2026) before implementation
+- `.claude/agents/commit-manager.md` - Manages git commits and workflow
+- `.claude/agents/claude-md-compactor.md` - Compacts CLAUDE.md when > 40k chars
+- `.claude/agents/tester-unit.md` - Creates unit tests with Vitest
+### Archived Agents
+- `.claude/agents/_archive/` - 82+ archived agents (unused, kept for reference)
+### Skills (20 active)
+- `.claude/skills/bun-runtime/` - Bun runtime patterns
+- `.claude/skills/codebase-knowledge/` - Domain mapping system
+- `.claude/skills/debugging-patterns/` - Error resolution patterns
+- `.claude/skills/docker-patterns/` - Docker multi-stage and security
+- `.claude/skills/docs-tracker/` - Documentation tracking
+- `.claude/skills/final-check/` - Final validation rules
+- `.claude/skills/git-workflow/` - Git workflow rules
+- `.claude/skills/nextjs-app-router/` - Next.js App Router patterns
+- `.claude/skills/performance-patterns/` - Performance profiling
+- `.claude/skills/quality-gate/` - Quality gate definitions
+- `.claude/skills/react-patterns/` - React 19 patterns
+- `.claude/skills/research-cache/` - Best practices research cache
+- `.claude/skills/security-scan/` - Security audit rules (OWASP)
+- `.claude/skills/shadcn-ui/` - shadcn/ui theming and accessibility
+- `.claude/skills/tailwind-patterns/` - Tailwind CSS patterns
+- `.claude/skills/test-coverage/` - Test coverage tracking
+- `.claude/skills/trpc-api/` - tRPC type-safe API patterns
+- `.claude/skills/typescript-strict/` - TypeScript strict mode
+- `.claude/skills/ui-ux-audit/` - UI/UX audit rules
+- `.claude/skills/zod-validation/` - Zod validation schemas
+### Plugins (5 via enabledPlugins)
+- `typescript-lsp` - Type diagnostics, go-to-def (LSP server, auto)
+- `code-review` - Code quality, PR analysis (/code-review cmd)
+- `security-guidance` - OWASP, vulnerability scan (PreToolUse hook, auto)
+- `commit-commands` - Git commit, push, PR flows (/commit cmd)
+- `frontend-design` - Production-grade UI design (/frontend-design cmd)
+### Configuration
+- `.claude/settings.json` - Agent registration, flows, rules, plugins
+- `.claude/CLAUDE.md` - System documentation (agents, skills, architecture)
+- `.claude/config/project-config.json` - Stack, structure, commands
+- `.claude/config/quality-gates.json` - Quality check commands
+- `.claude/config/testing-config.json` - Test framework config
+- `.claude/config/security-rules.json` - Security audit rules
+### Scripts
+- `.claude/scripts/validate-skills.sh` - Skills activation check
+- `.claude/scripts/mcp-quick-install.ts` - MCP server installer
+- `.claude/scripts/setup-mcps.ts` - MCP setup automation
+### Root Files
+- `CLAUDE.md` - Project rules and conventions
+- `.claude/CLAUDE.md` - Detailed system architecture
+## Connections
+- **All domains:** Documentation created on request only (not auto-generated)
+- **Skills:** Auto-injected into context when task matches description
+## Recent Commits
+| Hash    | Date       | Description                                                           |
+| ------- | ---------- | --------------------------------------------------------------------- |
+| pending | 2026-02-28 | refactor: remove hooks, update architecture to 6 agents + 20 skills   |
+| 6f3d9ff | 2026-02-24 | docs(skills): add skills activation research plan                     |
+| efa7c4f | 2026-02-24 | refactor(skills): rewrite all 23 descriptions to imperative pattern   |
+## Attention Points
+- [2026-02-28] **No hooks** - Hooks (stop-validator, user-prompt-submit) were removed. No automated enforcement.
+- [2026-02-28] **20 skills** - Down from 23. Removed: hook-development, mongoose-patterns, playwright-automation
+- [2026-02-24] **Imperative descriptions** - All skill descriptions use "ALWAYS invoke when... Do NOT..." pattern for 95%+ auto-invocation
+- [2026-02-19] **6 agents + 5 plugins** - Agents are subagents. Plugins auto-install via enabledPlugins in settings.json
+- [2024-12-19] **Config files for project specifics** - Agents read from `.claude/config/*.json` instead of hardcoding
+- [2024-12-19] **SKILL.md YAML frontmatter** - All skills MUST have `---` frontmatter with name, description