cursor-kenji

A curated toolkit of Cursor AI Agent Skills, Commands, MCP configs, Subagents, and Rules
Designed for modern full-stack development with React, Next.js, Supabase, and Tailwind

Living repository — skills evolve with the ecosystem. Every update is versioned.

How It All Fits Together

graph TB
  subgraph TOOLKIT["cursor-kenji Toolkit"]
    direction TB
    SK["Skills (60)"]
    CS["Cursor Skills (12)"]
    CMD["Commands (13)"]
    SA["Subagents (5)"]
    MCP["MCP Servers (16)"]
    RULES["Project Rules (8)"]
    NP["Notepads (2)"]
  end

  CURSOR["Cursor IDE"]
  YOU["You"]

  YOU -->|"types /command"| CMD
  YOU -->|"describes task"| SK
  YOU -->|"triggers keyword"| SA
  CURSOR -->|"auto-selects"| SK
  CURSOR -->|"auto-delegates"| SA
  CMD -->|"orchestrates"| SK
  SK -->|"calls"| MCP
  SA -->|"uses"| SK
  SA -->|"calls"| MCP
  RULES -->|"guides"| CURSOR
  NP -->|"provides context"| CURSOR
  CS -->|"extends"| CURSOR

  style TOOLKIT fill:#1e1b4b,stroke:#6366f1,stroke-width:2px,color:#e0e7ff
  style CURSOR fill:#0f172a,stroke:#38bdf8,stroke-width:2px,color:#e0f2fe
  style YOU fill:#064e3b,stroke:#10b981,stroke-width:2px,color:#d1fae5
  style SK fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style CS fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style CMD fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style SA fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style MCP fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style RULES fill:#3b0764,stroke:#a78bfa,color:#ede9fe
  style NP fill:#3b0764,stroke:#a78bfa,color:#ede9fe

What's New in 2026

Based on Cursor's official agent best practices and the latest Cursor features (Jan-May 2026).

May 2026 — Native RN Loop & Cross-Surface UI ^Latest

Two skills + two rules from real RN/Capacitor + Supabase shipping sessions — the "no-Mac dev loop" and the "looks great on web, atrocious on phone" patterns.

Addition	Type	Why It Matters
`start-emulator`	Skill	Boots Metro + Android emulator in the right order — kills duplicate ports, picks fresh-cache vs fast-iteration, defaults to 1080×4000 for tall QA screenshots, polls `/status` before deeplink to avoid "Cannot connect to Expo CLI" races. Pairs with `test-emulator`
`enhance-web-mobile-ui`	Skill	Architecture skill for hybrid web/iOS/Android apps (Capacitor / Tauri / Expo Web / Ionic). Establishes three orthogonal axes — form factor, platform, pointer capability — so a UIUX sweep on one surface cannot silently degrade another. Catches axis conflation, viewport-as-form-factor abuse, hover-only affordances on touch shells
`full-stack-ship-discipline.mdc`	Rule	`alwaysApply: true` — prevents the "local migration file never deployed" failure mode. Forces UI tasks to inventory backend deps (schema, RPC, RLS, edge functions) in the same chat, deploy via Supabase MCP, and verify against the remote DB
`shell-first-search.md`	Rule	Workspace-wide rule routing routine search to `Shell` (`grep`/`find`/`ls`) instead of the `Grep`/`Glob` tools, which have hung for minutes on this Windows host. Keeps `SemanticSearch` and `Read` as the correct tools for their use cases

May 2026 — Housekeep & Direct-Tone Pass

Library housekeep aligning with Cursor's official skill spec and direct-tone description guidance.

Change	Type	Why It Matters
11 skills renamed	Skill	`name:` frontmatter now matches folder names per Cursor spec — auto-loader was silently failing on `accessibility-audit`, `code-review`, `performance-audit`, `security-audit`, `api-design`, `frontend-design`, `doc-coauthoring`, `mcp-builder`, `skill-creator`, `git-workflow`, `refactoring`
22 descriptions tightened	Skill	Verbose `description: >` blocks (some 280+ words) reduced to direct-tone single-sentence WHAT + concrete trigger list, matching Cursor's canonical example pattern
9 commands demoted to pointers	Command	`/commit`, `/debug`, `/pr`, `/readme`, `/refactor`, `/review`, `/test`, `/uiux`, `/update-deps` now point to their skill equivalents instead of duplicating the playbook. Skills handle natural language; commands handle explicit slash invocation
`audit-ux` cross-reference	Skill	Fixed broken pointer (`audit-ui-design-system` → `audit-uiux-design-system`)

Apr 2026 — The Enhance Family

Generic, research-grounded skills that turn AI-templated screens into hand-crafted ones, applicable to any web stack.

Addition	Type	Why It Matters
`enhance-page-ui`	Skill	Composition over decoration — fix hierarchy, grouping, spacing, motion before adding visual flourish. NN/g visual hierarchy + Laws of UX grounded
`enhance-page-ux`	Skill	Replace generic "stacked" UI with semantic data — every change cites a Nielsen heuristic, uses existing primitives, verified at 3 viewports via browser MCP
`enhance-readme`	Skill	Theme-aware hero + tour grid + animated GIF — turns plain READMEs into showcases via Playwright MCP screenshot capture
`audit-ux`	Skill	Deep UX audit grounded in NN/g 10 heuristics, Laws of UX, Intuit Content Design, and Google HEART — generic across any webapp
`split-to-prs`	Cursor Skill	Slice a single chat / branch / PR into small reviewable PRs with safe snapshot, no destructive git ops
Updated `canvas`	Cursor Skill	Live React canvas SDK — refreshed primitives, better SDK type hints, additional design guidance
Updated `/commit`	Command	Lint, Sentry pre-check, build verify, auto-detect scope, conventional commit, push — full pre-commit pipeline
Updated `/research`	Command	Three-phase Firecrawl deep research with gap analysis, fallback to WebSearch, sequential thinking for complex changes
Updated `/readme`	Command	End-of-session doc sync — detect convention, smart change detection, stale reference cleanup
Updated `/review`, `/fix-issue`, `/update-deps`	Commands	More thorough, safer, more actionable

Earlier 2026

Addition	Type	Why It Matters
`hooks-builder`	Skill	Cursor Hooks — auto-format on edit, block dangerous commands, scan for secrets, create agent loop automation
`tdd`	Skill	TDD is Cursor's #1 recommended agent pattern — tests give agents a clear, verifiable goal
`spec-writing`	Skill	Writing good specs is the highest-leverage 2026 AI skill — vague prompts produce vague code
`parallel-agents`	Skill	Worktrees + cloud agents + multi-model comparison — delegate and compare in parallel
`audit-code-review`	Skill	Agent Review + BugBot + manual checklist for thorough pre-merge review
20 new skills	Skill	Audits, debugging, deploy verification, file handling, PRD generation, QA testing, housekeeping, and more
6 new cursor-skills	Cursor	babysit, canvas, create-hook, shell, statusline, update-cli-config
`/plan`	Command	Plan Mode (`Shift+Tab`) — Cursor's #1 recommendation: plan before coding
`/pr`	Command	Checks pass -> commit -> push -> open PR with description, one workflow
`/debug`	Command	Debug Mode — hypothesis-driven, instruments code, pinpoints root cause

What's Inside

pie title Toolkit Composition
    "Skills (60)" : 60
    "Commands (13)" : 13
    "MCP Servers (16)" : 16
    "Project Rules (8)" : 8
    "Subagents (5)" : 5
    "Cursor Skills (12)" : 12
    "Notepads (2)" : 2
    "Shell Aliases (8)" : 8

Core

	Count	Description
	Skills	60
	Cursor Skills	12
	Commands	13
	Subagents	5

Configuration

	Count	Description
	MCP Servers	16
	Project Rules	8
	Notepads	2
	Shell Aliases	8

Full directory tree

cursor-kenji/
├── skills/                  # 58 Agent Skills (each has SKILL.md)
│   ├── algorithmic-art/
│   ├── audit-accessibility/
│   ├── audit-code-review/
│   ├── audit-db-schema/
│   ├── audit-fe-api/
│   ├── audit-langfuse-llm/
│   ├── audit-performance/
│   ├── audit-security/
│   ├── audit-uiux-design-system/
│   ├── audit-ux/              # NEW Apr 2026 — NN/g + Laws of UX + HEART audit
│   ├── backend-patterns/
│   ├── canvas-design/
│   ├── code-antipatterns/
│   ├── codebase-coherency/
│   ├── creative-effects/
│   ├── creative-workflow/
│   ├── data-visualization/
│   ├── database-optimization/
│   ├── debug-error/
│   ├── debug-fe-be-integration/
│   ├── debug-sentry-monitor/
│   ├── deploy-verify/
│   ├── design-api/
│   ├── design-frontend/
│   ├── design-prd/
│   ├── design-system/
│   ├── docs-coauthor/
│   ├── docs-writer/
│   ├── enhance-page-ui/       # composition + hierarchy + motion
│   ├── enhance-page-ux/       # heuristic-grounded UX enhancement
│   ├── enhance-readme/        # hero + tour + GIF for any README
│   ├── enhance-web-mobile-ui/ # NEW May 2026 — cross-surface (web/iOS/Android) architecture
│   ├── error-handling/
│   ├── file-docx/
│   ├── file-pdf/
│   ├── file-pptx/
│   ├── file-xlsx/
│   ├── hooks-builder/
│   ├── interactive-ux/
│   ├── karpathy-guidelines/
│   ├── meta-mcp-builder/
│   ├── meta-skill-creator/
│   ├── mobile-first/
│   ├── motion-design/
│   ├── parallel-agents/
│   ├── protocol-browser-anti-stall/
│   ├── realtime-features/
│   ├── spec-writing/
│   ├── start-emulator/        # NEW May 2026 — Metro + Android emulator bring-up
│   ├── tdd/
│   ├── test-emulator/
│   ├── test-qa/
│   ├── test-unit/
│   ├── theme-factory/
│   ├── uiux-enhancement/
│   ├── webapp-testing/
│   ├── workflow-git-commit/
│   ├── workflow-housekeep/
│   ├── workflow-pr/
│   └── workflow-refactor/
├── skills-cursor/           # 12 Cursor-specific Skills
│   ├── babysit/
│   ├── canvas/                # Updated Apr 2026 — refreshed SDK + design rules
│   ├── create-hook/
│   ├── create-rule/
│   ├── create-skill/
│   ├── create-subagent/
│   ├── migrate-to-skills/
│   ├── shell/
│   ├── split-to-prs/          # NEW Apr 2026 — safely slice work into small PRs
│   ├── statusline/
│   ├── update-cli-config/
│   └── update-cursor-settings/
├── commands/                # 13 Slash Commands
│   ├── commit.md
│   ├── debug.md
│   ├── fix-issue.md
│   ├── mcp.md
│   ├── plan.md
│   ├── pr.md
│   ├── readme.md
│   ├── refactor.md
│   ├── research.md
│   ├── review.md
│   ├── test.md
│   ├── uiux.md
│   └── update-deps.md
├── agents/                  # 5 Subagents
│   ├── code-reviewer.md
│   ├── db-migrator.md
│   ├── debugger.md
│   ├── deploy-checker.md
│   └── perf-monitor.md
├── rules/
│   ├── senior-engineer.md           # Global rule — full-stack execution protocol
│   ├── full-stack-ship-discipline.mdc  # NEW May 2026 — alwaysApply, migrations must deploy
│   ├── shell-first-search.md        # NEW May 2026 — route search to Shell, not Grep/Glob
│   ├── project-starter/             # 6 Project rule templates
│   └── native-rn-monorepo/          # RN + Web monorepo, CI-only iOS bundle
│       ├── components.mdc
│       ├── data-fetching.mdc
│       ├── git.mdc
│       ├── supabase.mdc
│       ├── tailwind.mdc
│       └── typescript.mdc
├── notepads/                # Reusable context templates
│   ├── architecture.md
│   └── design-tokens.md
├── shell-aliases/
│   └── cursor-helpers.sh    # 8 shell commands
├── mcp/                     # MCP server configs
│   ├── README.md
│   ├── mcp.json.template      (Essential 5 servers)
│   └── mcp-full.json.template (All 16 servers)
├── docs/
│   ├── CATALOG.md
│   └── CONTRIBUTING.md
├── install.sh               # One-command installer
├── LICENSE
└── README.md

Quick Start

git clone https://github.com/kensaurus/cursor-kenji.git
cd cursor-kenji
./install.sh

Or one-liner install

curl -sSL https://raw.githubusercontent.com/kensaurus/cursor-kenji/main/install.sh | bash

What the installer does

flowchart LR
  A["./install.sh"] --> B["Backup existing\n~/.cursor/skills/"]
  B --> C["Install 57 skills\n+ 12 cursor skills"]
  C --> D["Install 5\nsubagents"]
  D --> E["Copy MCP\nconfig template"]
  E --> F["Ready!"]

  style A fill:#064e3b,stroke:#10b981,stroke-width:2px,color:#d1fae5
  style F fill:#064e3b,stroke:#10b981,stroke-width:2px,color:#d1fae5
  style B fill:#1e293b,stroke:#64748b,color:#e2e8f0
  style C fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style D fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style E fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe

Post-install steps:

Restart Cursor to pick up new skills
Edit ~/.cursor/mcp.json with your API keys
Try: /commit, /test, /research in Cursor
Source shell helpers: source ~/cursor-kenji/shell-aliases/cursor-helpers.sh

How to Use This Toolkit

Five primitives, one mental model: describe the task → Cursor auto-selects a skill, OR type a /command for a known workflow.

The four ways the agent gets help

flowchart LR
  YOU["You"] -->|"natural language\n('clean up this page')"| AUTO["Skills auto-trigger\nfrom description"]
  YOU -->|"slash command\n('/commit')"| CMD["Commands\nrun a recipe"]
  YOU -->|"keyword in chat\n('migration', 'review')"| SUB["Subagents\nauto-delegate"]
  YOU -->|"long-lived guidance\n(.cursor/rules/)"| RULES["Project Rules\nalways apply"]

  AUTO --> WORK["Cursor Agent\ndoes the work"]
  CMD --> WORK
  SUB --> WORK
  RULES -.->|"shape every response"| WORK

  style YOU fill:#064e3b,stroke:#10b981,color:#d1fae5
  style WORK fill:#1e1b4b,stroke:#6366f1,color:#e0e7ff
  style AUTO fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style CMD fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style SUB fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style RULES fill:#3b0764,stroke:#a78bfa,color:#ede9fe

Primitive	When	How to invoke	Example
Skill	Specific task with a known workflow (audit, enhance, debug, build)	Just describe the task — Cursor matches keywords from skill descriptions	"make `/dashboard` feel less AI-generated" → triggers `enhance-page-ux`
Command	Repeatable workflow you run often	Type `/<name>` in chat	`/commit`, `/research`, `/pr`
Subagent	Background autonomous work (review, migrate, deploy)	Mention a trigger keyword — Cursor auto-delegates	"review this PR" → `code-reviewer`
Rule	Always-on project conventions	Drop `.mdc` into `.cursor/rules/` of any project	`supabase.mdc`, `typescript.mdc`

Common daily flows

flowchart TB
  subgraph FEAT["Building a feature"]
    F1["/research <topic>"] --> F2["/plan"]
    F2 --> F3["Code"]
    F3 --> F4["/test"]
    F4 --> F5["/commit"]
    F5 --> F6["/pr"]
  end

  subgraph FIX["Fixing a bug"]
    B1["/debug or /fix-issue #N"] --> B2["Code"]
    B2 --> B3["/review"]
    B3 --> B4["/commit"]
  end

  subgraph POLISH["Polishing UI"]
    P1["audit-ux skill\n(report)"] --> P2["enhance-page-ux skill\n(implement)"]
    P2 --> P3["enhance-page-ui skill\n(visual polish)"]
    P3 --> P4["/commit"]
  end

  subgraph DOCS["Shipping docs"]
    D1["enhance-readme skill"] --> D2["/readme"]
    D2 --> D3["/commit"]
  end

  style FEAT fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style FIX fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style POLISH fill:#3b0764,stroke:#a78bfa,color:#ede9fe
  style DOCS fill:#064e3b,stroke:#10b981,color:#d1fae5

How to invoke a skill (without overthinking it)

Skills auto-trigger from trigger phrases in their description (the description: block at the top of every SKILL.md). You don't need to remember names — just describe what you want.

You say	Skill that fires
"this page feels AI-generated, fix it"	`enhance-page-ux`
"make `/settings` more polished, less crowded"	`enhance-page-ui`
"give the README a hero image and screenshots"	`enhance-readme`
"audit the UX of the checkout flow"	`audit-ux`
"split this branch into smaller PRs"	`split-to-prs`
"show me a canvas with these query results"	`canvas`
"the agent keeps hanging on browser steps"	`protocol-browser-anti-stall`

If you want to force a skill, mention it explicitly: "use enhance-page-ux on /dashboard".

How the new "Enhance" family works

The three enhance-* skills compose like a pipeline — pick the layer that matches what you want changed:

flowchart LR
  AUDIT["audit-ux\n(diagnose)"] -.->|"report only"| REPORT["UX report"]
  AUDIT --> UX
  UX["enhance-page-ux\n(workflow + data)"] --> UI
  UI["enhance-page-ui\n(composition + motion)"] --> DONE["Polished page"]
  README["enhance-readme\n(showcase the repo itself)"] --> SHIP["Pretty README"]

  style AUDIT fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style UX fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style UI fill:#3b0764,stroke:#a78bfa,color:#ede9fe
  style README fill:#064e3b,stroke:#10b981,color:#d1fae5
  style DONE fill:#1e1b4b,stroke:#6366f1,color:#e0e7ff
  style SHIP fill:#1e1b4b,stroke:#6366f1,color:#e0e7ff

Want to...	Use
Get a heuristic-grounded report (NN/g + Laws of UX + HEART, no code changes)	`audit-ux`
Replace stacked / templated UI with semantic data that maps to real backend state	`enhance-page-ux`
Refine layout, hierarchy, spacing, motion of an already-correct page	`enhance-page-ui`
Visually showcase the repo itself with hero image + tour + animated GIF	`enhance-readme`

All four are generic — they work in any web stack (Next.js, Remix, SvelteKit, Vite + React, etc.) because they rely on browser MCP for live observation and on inventory of existing primitives rather than assumed framework APIs.

Setting up a new project

Drop project rules — copy the .mdc files into the project's .cursor/rules/:

mkdir -p .cursor/rules
cp ~/cursor-kenji/rules/project-starter/*.mdc .cursor/rules/

Add an AGENTS.md at the repo root if you want the agent to follow specific conventions across the whole repo (Cursor reads it automatically).
Verify MCP keys in ~/.cursor/mcp.json — Firecrawl, Supabase, GitHub PAT — only what you actually use.
Try one workflow end-to-end: /research "current React Server Components patterns" → /plan → code → /test → /commit → /pr. If anything is missing, run cursor-sync to refresh.

Tips for getting the most out of skills

Be specific about scope. "Enhance the page" is vague. "Enhance /dashboard so empty cells get folder-aggregate counts and the toolbar buttons stop wrapping at 1024px" gives the agent a concrete target.
Let skills cite heuristics. audit-ux and enhance-page-ux insist every change names the heuristic it satisfies — review the citation, not just the diff. If the citation is weak, the change probably is too.
Stack a slow-then-fast workflow. /plan first (slow, in Plan Mode) → switch to Agent Mode for execution → /review before /commit → /pr at the end.
Compose skills. Run audit-ux → review report → run enhance-page-ux for the top 3 issues → run enhance-page-ui for visual polish → /commit. Each skill leaves clear write-up tables for the next one.
Don't fight the rules. Project rules in .cursor/rules/*.mdc constrain the agent; if a rule blocks something legitimate, edit the rule rather than ignoring it.
Update often. cd ~/cursor-kenji && git pull && ./install.sh.

Skills (60)

Skill Categories at a Glance

mindmap
  root((60 Skills))
    Enhance
      enhance-page-ui
      enhance-page-ux
      enhance-readme
      enhance-web-mobile-ui
    Design & Frontend
      design-frontend
      design-system
      motion-design
      creative-effects
      uiux-enhancement
      interactive-ux
      mobile-first
      theme-factory
    Data & Creative
      data-visualization
      algorithmic-art
      canvas-design
    Backend & Database
      backend-patterns
      database-optimization
      realtime-features
    Architecture & Quality
      design-api
      error-handling
      code-antipatterns
      audit-code-review
      codebase-coherency
      workflow-refactor
      audit-performance
      audit-security
      audit-accessibility
    Audits & Monitoring
      audit-db-schema
      audit-fe-api
      audit-langfuse-llm
      audit-uiux-design-system
      audit-ux
      debug-sentry-monitor
      deploy-verify
    Debugging
      debug-error
      debug-fe-be-integration
    Engineering Practices
      tdd
      spec-writing
      parallel-agents
      hooks-builder
      karpathy-guidelines
      test-unit
      test-qa
      test-emulator
      start-emulator
      workflow-pr
      protocol-browser-anti-stall
    Process & Docs
      workflow-git-commit
      docs-coauthor
      docs-writer
      design-prd
      creative-workflow
      workflow-housekeep
    File Handling
      file-docx
      file-pdf
      file-pptx
      file-xlsx
    Meta & Tooling
      meta-skill-creator
      meta-mcp-builder
      webapp-testing

Enhance ^{New Apr-May 2026}

Make existing pages and READMEs feel hand-crafted. Generic across any web stack.

Skill	What it Does
`enhance-page-ui`	Composition before decoration — hierarchy, grouping, spacing, motion. Subtract clutter, group related, pin metadata, soften scroll edges. NN/g + Laws of UX grounded
`enhance-page-ux`	Replace stacked / templated UI with semantic data wired to real backend state. Every change cites a Nielsen heuristic, uses existing primitives, verified at 1440/1024/800 viewports
`enhance-readme`	Theme-aware hero (`<picture>` dark/light auto-swap) + tour grid + optional autoplay GIF via Playwright MCP. Works for any repo with a live URL or local dev server
`enhance-web-mobile-ui` ^{May 2026}	Cross-surface architecture for hybrid PWA + iOS + Android apps (Capacitor / Tauri / Expo Web / Ionic). Three orthogonal axes (form factor / platform / pointer) + mode tokens + container-query primitives — so polish on one surface can't degrade another

Design & Frontend

Build distinctive, production-grade interfaces

Skill	What it Does
`design-frontend`	Production-grade UI avoiding generic AI aesthetics
`design-system`	Component libraries, tokens, variants, CVA patterns
`motion-design`	Framer Motion, CSS animations, GSAP micro-interactions
`creative-effects`	WebGL, Three.js, shaders, particles, Canvas 2D
`uiux-enhancement`	Incremental UI/UX improvements and polish
`interactive-ux`	Gamification, Easter eggs, delightful interactions
`mobile-first`	Touch-optimized, responsive, PWA patterns
`theme-factory`	Apply cohesive visual themes across artifacts

Data & Creative

Skill	What it Does
`data-visualization`	Recharts, D3.js, sparklines, real-time charts
`algorithmic-art`	Generative art, flow fields, L-systems, circle packing
`canvas-design`	Museum-quality visual design in `.png` and `.pdf` formats

Backend & Database

Skill	What it Does
`backend-patterns`	Server Actions, tRPC, Edge Functions, caching, jobs
`database-optimization`	Indexes, N+1 fixes, RLS performance, query tuning
`realtime-features`	WebSocket, Supabase Realtime, SSE, live data

Architecture & Quality

Skill	What it Does
`design-api`	REST conventions, error schemas, pagination, versioning
`error-handling`	Error boundaries, Server Action errors, toast patterns
`code-antipatterns`	Detect and fix React, TypeScript, state anti-patterns
`audit-code-review`	Thorough PR reviews — correctness, security, perf, a11y checklist
`codebase-coherency`	Naming, imports, organization consistency audit
`workflow-refactor`	Safe, incremental code transformations
`audit-performance`	Core Web Vitals, bundle analysis, runtime profiling
`audit-security`	OWASP Top 10, auth flows, RLS, secrets management
`audit-accessibility`	WCAG 2.1 AA compliance, screen reader, keyboard, ARIA

Engineering Practices ^{New in 2026}

Skill	What it Does
`tdd`	Test-driven development with AI — Red/Green/Refactor, Vitest patterns, agent-compatible TDD workflow
`spec-writing`	Write effective specs and briefs so agents produce correct implementations first time
`parallel-agents`	Run agents in parallel via git worktrees, cloud agents, and multi-model comparison
`hooks-builder`	Build Cursor Agent Hooks — auto-formatters, security gates, secret scanners, loop automation
`karpathy-guidelines`	Behavioral guardrails (Think before coding, Simplicity first, Surgical changes, Goal-driven execution) — distilled from Karpathy's LLM coding pitfalls

Process & Documentation

Skill	What it Does
`workflow-git-commit`	Branching, conventional commits, PRs, releases
`docs-coauthor`	Structured co-authoring for specs, PRDs, RFCs
`creative-workflow`	End-to-end feature development workflow

Audits & Monitoring ^New

Deep, MCP-powered audits across the full stack

Skill	What it Does
`audit-db-schema`	Database schema audit — naming, types, constraints, indexes, RLS, migrations, security
`audit-fe-api`	Frontend API calls vs backend — contract alignment, caching, error handling
`audit-langfuse-llm`	PDCA audit for LLM features — traces, prompts, costs, evals, grounding via Langfuse
`audit-uiux-design-system`	Visual token compliance vs design system — colors, spacing, components, WCAG
`audit-ux` ^NEW	Generic UX audit — NN/g 10 heuristics, Laws of UX, Intuit Content Design, HEART metrics. Stack-agnostic
`debug-sentry-monitor`	Sentry issue triage, root cause analysis, noise filtering, architecture audit
`deploy-verify`	Post-deploy smoke test — Sentry + Supabase + Langfuse + Playwright, ship-or-rollback verdict

Debugging ^New

Systematic root cause analysis, not guessing

Skill	What it Does
`debug-error`	Systematic debugging workflow — reproduce, isolate, research, fix, verify, prevent
`debug-fe-be-integration`	FE/BE integration debug — backend logs, API mismatches, both-side fixes

Testing & QA ^New

From unit tests to full E2E QA

Skill	What it Does
`test-unit`	Auto-detect framework, research patterns, Sentry coverage gaps, write tests
`test-qa`	Comprehensive QA via browser MCP — CRUD lifecycle, data pipeline, UX quality audit
`test-emulator`	Native build QA on Android emulator — Metro/adb walk + Supabase + Sentry MCPs, three-layer CRUD verification, fixes for white-screen / cache-rehydration / sync-empty-state failure modes
`start-emulator` ^{NEW May 2026}	Boot Metro + Android emulator in the correct order — kills duplicate ports, fresh-cache vs fast-iteration choice, polls `/status` before deeplink to avoid "Cannot connect to Expo CLI" races. Pairs with `test-emulator`
`workflow-pr`	PR lifecycle — validation, monitoring, bot feedback, merge criteria
`protocol-browser-anti-stall`	Anti-stall protocol for browser automation — timeouts, retries, evidence gathering

Product & Documentation ^New

From PRD to production docs

Skill	What it Does
`design-prd`	Generate PRDs via structured conversation — competitive research, technical feasibility
`docs-writer`	Write READMEs, API docs, architecture docs, code comments
`workflow-housekeep`	Full-cycle repo maintenance — README sync, dead file cleanup, dependency updates, config audit

File Handling ^New

Create and manipulate Office documents and PDFs

Skill	What it Does
`file-docx`	Word documents — create, edit, tracked changes, comments, text extraction
`file-pdf`	PDF processing — extract text/tables, create, merge/split, forms, OCR
`file-pptx`	PowerPoint — create from HTML, edit slides, extract content, visual validation
`file-xlsx`	Spreadsheets — formulas, formatting, data analysis, financial model standards

Meta & Tooling

Skill	What it Does
`meta-skill-creator`	Guide for creating new Agent Skills
`meta-mcp-builder`	Build MCP servers for LLM tool integration
`webapp-testing`	Playwright browser automation and E2E testing

Cursor-Specific Skills (12)

Skill	What it Does
`babysit`	Keep a PR merge-ready — triage comments, resolve conflicts, fix CI in a loop
`canvas`	Live React canvas beside chat — rich data visualizations, audit reports, interactive tools
`create-hook`	Create Cursor hooks — scripts/prompts for before/after agent events
`create-rule`	Create `.cursor/rules/` for persistent AI guidance
`create-skill`	Create new Agent Skills in `~/.cursor/skills/`
`create-subagent`	Create custom subagents in `.cursor/agents/`
`migrate-to-skills`	Convert rules/commands to Skills format
`shell`	Direct shell execution — run `/shell` commands without interpretation
`split-to-prs` ^NEW	Slice one pile of work into small reviewable PRs — safe snapshot, no destructive git ops, approval-gated
`statusline`	Configure CLI status line — model, context usage, git info
`update-cli-config`	Modify CLI settings — permissions, sandbox, vim mode, display
`update-cursor-settings`	Modify Cursor/VSCode settings.json

Commands (13)

Development Workflow

flowchart LR
  PLAN["/plan"] --> CODE["Write Code"]
  CODE --> TEST["/test"]
  TEST --> DEBUG{Bugs?}
  DEBUG -->|Yes| FIX["/debug"]
  FIX --> TEST
  DEBUG -->|No| REVIEW["/review"]
  REVIEW --> COMMIT["/commit"]
  COMMIT --> PR["/pr"]

  style PLAN fill:#064e3b,stroke:#10b981,color:#d1fae5
  style CODE fill:#1e293b,stroke:#64748b,color:#e2e8f0
  style TEST fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style DEBUG fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style FIX fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style REVIEW fill:#3b0764,stroke:#a78bfa,color:#ede9fe
  style COMMIT fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style PR fill:#064e3b,stroke:#10b981,color:#d1fae5

Coding Workflow

Command	When	What it Does
`/plan`	Before coding	Plan Mode — research codebase, clarify requirements, produce an approved plan before writing code
`/commit`	After coding	Fix build errors, lint, type check, commit & push
`/pr`	Ready to ship	Checks pass -> commit -> push -> open PR with title and description
`/fix-issue [#]`	Bug reports	Fetch GitHub issue -> find relevant code -> implement fix -> open PR
`/debug`	Tricky bugs	Hypothesis-driven debugging with runtime instrumentation, not guessing
`/review`	Before merge	Agent review pass + manual checklist: correctness, security, performance, accessibility
`/test`	Before commit	Run full test suite, verify quality, check coverage targets
`/update-deps`	Maintenance	Audit and safely update dependencies one at a time with changelog review

Research & Documentation

Command	When	What it Does
`/research`	Before coding	Scrape latest docs, patterns, and solutions via Firecrawl
`/readme`	End of session	Sync all READMEs to reflect session changes
`/refactor`	Long files	Split into clean, modular architecture without losing any code
`/mcp`	Dev workflow	MCP-powered development reference and tool guide
`/uiux`	UI review	Enforce design system, fix rogue styling, standardize interactions

Bundles

Drop-in command + rule bundles for specific project shapes

Bundle	Path	What it ships
`native-rn-monorepo`	`commands/native-rn-monorepo/` `rules/native-rn-monorepo/`	9 commands (`/android-`, `/ios-ci-`, `/rn-*`) + 5 rules for an RN + Web monorepo where the dev is on Linux/Windows and iOS verification is CI-only

# Install the native-rn-monorepo bundle into a project
mkdir -p <project>/.cursor/{commands,rules}
cp ~/cursor-kenji/commands/native-rn-monorepo/*.md  <project>/.cursor/commands/
cp ~/cursor-kenji/rules/native-rn-monorepo/*.mdc    <project>/.cursor/rules/

Subagents (5)

Autonomous AI agents that Cursor auto-delegates to based on keywords

flowchart TB
  CURSOR["Cursor IDE"] -->|"code changes"| CR["code-reviewer"]
  CURSOR -->|"errors / exceptions"| DB["debugger"]
  CURSOR -->|"migration / new table"| DM["db-migrator"]
  CURSOR -->|"deploy / ship it"| DC["deploy-checker"]
  CURSOR -->|"slow / optimize"| PM["perf-monitor"]

  CR -->|output| QR["Quality Report\nSecurity + Types + Anti-patterns"]
  DB -->|output| RCA["Root Cause Analysis\nIsolate + Fix + Verify"]
  DM -->|output| MIG["SQL Migration\nRLS + Indexes + Types"]
  DC -->|output| CHK["8-Point Checklist\nPre-deploy Validation"]
  PM -->|output| PERF["Performance Report\nBundle + Render + Data"]

  style CURSOR fill:#0f172a,stroke:#38bdf8,stroke-width:2px,color:#e0f2fe
  style CR fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style DB fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style DM fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style DC fill:#064e3b,stroke:#10b981,color:#d1fae5
  style PM fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style QR fill:#1e1b4b,stroke:#6366f1,color:#e0e7ff
  style RCA fill:#450a0a,stroke:#ef4444,color:#fee2e2
  style MIG fill:#0c2340,stroke:#3b82f6,color:#dbeafe
  style CHK fill:#022c22,stroke:#059669,color:#d1fae5
  style PERF fill:#451a03,stroke:#d97706,color:#fef3c7

Agent	Auto-triggers On	What it Does
`code-reviewer`	Code changes, "review"	Quality, security, types, anti-patterns
`debugger`	Errors, exceptions	Root cause analysis, isolate, fix, verify
`db-migrator`	"migration", "new table"	SQL, RLS policies, indexes, type generation
`deploy-checker`	"deploy", "ship it"	8-check validation pipeline
`perf-monitor`	"slow", "optimize"	Bundle, render, data fetching audit

MCP Servers (16)

External tool integrations across 4 tiers — pick what you need

graph TB
  subgraph T1["Tier 1: Essential"]
    ST["Sequential Thinking"]
    C7["Context7"]
    FC["Firecrawl"]
    SB["Supabase"]
    CD["Chrome DevTools"]
  end

  subgraph T2["Tier 2: Dev Power-Ups"]
    GH["GitHub"]
    GHO["GitHub Official"]
    PW["Playwright"]
    PG["Postgres"]
    MEM["Memory"]
  end

  subgraph T3["Tier 3: AWS Cloud"]
    LAM["AWS Lambda"]
    S3["AWS S3"]
    CW["AWS CloudWatch"]
    RD["Redis"]
  end

  subgraph T4["Tier 4: Productivity"]
    SL["Slack"]
    NT["Notion"]
  end

  style T1 fill:#064e3b,stroke:#10b981,stroke-width:2px,color:#d1fae5
  style T2 fill:#1e3a5f,stroke:#60a5fa,stroke-width:2px,color:#dbeafe
  style T3 fill:#78350f,stroke:#fbbf24,stroke-width:2px,color:#fef3c7
  style T4 fill:#3b0764,stroke:#a78bfa,stroke-width:2px,color:#ede9fe

Tier 1: Essential

Server	Key?
Sequential Thinking	No
Context7	No
Firecrawl	Yes
Supabase	Yes
Chrome DevTools	No*

Tier 2: Dev Power-Ups

Server	Key?
GitHub	PAT
GitHub Official	PAT
Playwright	No
Postgres	Conn
Memory	No

Tier 3: AWS Cloud

Server	Key?
AWS Lambda	Profile
AWS S3	Profile
AWS CloudWatch	Profile
Redis	URL

Tier 4: Productivity

Server	Key?
Slack	Bot
Notion	Yes

Two templates included:

mcp.json.template — Essential 5 servers
mcp-full.json.template — All 16 servers

See mcp/README.md for setup guides.

Project Rules Starter Pack

Drop into any project's .cursor/rules/ for instant AI guidance

graph LR
  R1["supabase.mdc"] --> P["Your Project\n.cursor/rules/"]
  R2["components.mdc"] --> P
  R3["typescript.mdc"] --> P
  R4["tailwind.mdc"] --> P
  R5["git.mdc"] --> P
  R6["data-fetching.mdc"] --> P
  SE["senior-engineer.md\n(Global Rule)"] -.->|"alwaysApply: true"| P

  style P fill:#1e1b4b,stroke:#6366f1,stroke-width:2px,color:#e0e7ff
  style SE fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style R1 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style R2 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style R3 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style R4 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style R5 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style R6 fill:#064e3b,stroke:#10b981,color:#d1fae5

Rule	Enforces
`supabase.mdc`	Typed clients, RLS mandatory, migration patterns
`components.mdc`	Reuse primitives, Server Components, a11y
`typescript.mdc`	No `any`, Zod validation, ActionResult pattern
`tailwind.mdc`	Design tokens, `cn()`, mobile-first, motion prefs
`git.mdc`	Conventional commits, branch naming, no secrets
`data-fetching.mdc`	TanStack Query, prefetch, query key factories

Plus three global rules that always apply across any project:

Rule	Enforces
`senior-engineer.md`	Full-stack execution protocol with MCP tool usage
`full-stack-ship-discipline.mdc` ^{NEW May 2026}	Every UI task is full-stack until verified end-to-end — deploy migrations via Supabase MCP in the same chat, verify against the remote DB, not the local file
`shell-first-search.md` ^{NEW May 2026}	Route routine search to `Shell` (`grep`/`find`/`ls`) instead of the `Grep`/`Glob` tools, which can hang for minutes on some Windows hosts. `SemanticSearch` stays for meaning-based queries

cp ~/cursor-kenji/rules/project-starter/*.mdc your-project/.cursor/rules/
cp ~/cursor-kenji/rules/{senior-engineer.md,full-stack-ship-discipline.mdc,shell-first-search.md} your-project/.cursor/rules/

The native-rn-monorepo/ bundle is also available for React Native + Web monorepos targeting CI-only iOS — see rules/native-rn-monorepo/README.md.

Shell Helpers

source ~/cursor-kenji/shell-aliases/cursor-helpers.sh

Command	What it Does
`newskill <name>`	Create a new skill with template
`lsskills`	List all installed skills with descriptions
`cursor-sync`	Pull latest + reinstall
`cursor-dev`	Open Cursor + Chrome DevTools
`newrule <name>`	Create a project rule with template
`newagent <name>`	Create a subagent with template
`gc <type> <msg>`	Conventional commit shortcut
`gp`	Push current branch

Design Principles

graph LR
  P1["Check Existing\nFirst"] --> P2["Production\nReady"]
  P2 --> P3["Modular &\nComposable"]
  P3 --> P4["Creative Yet\nDisciplined"]
  P4 --> P5["Modern\nStack"]
  P5 --> P6["Accessible\nby Default"]
  P6 --> P7["Performance\nAware"]

  style P1 fill:#312e81,stroke:#818cf8,color:#e0e7ff
  style P2 fill:#064e3b,stroke:#10b981,color:#d1fae5
  style P3 fill:#1e3a5f,stroke:#60a5fa,color:#dbeafe
  style P4 fill:#78350f,stroke:#fbbf24,color:#fef3c7
  style P5 fill:#7f1d1d,stroke:#f87171,color:#fee2e2
  style P6 fill:#3b0764,stroke:#a78bfa,color:#ede9fe
  style P7 fill:#1e293b,stroke:#64748b,color:#e2e8f0

#	Principle
1	Check Existing First — scan before creating. Never duplicate.
2	Production-Ready — no placeholders. Code that ships.
3	Modular & Composable — skills cross-reference. Mix and match.
4	Creative Yet Disciplined — bold aesthetics, solid engineering.
5	Modern Stack — React 19, Next.js 15+, Tailwind v4, strict TS.
6	Accessible by Default — WCAG 2.1 AA is non-negotiable.
7	Performance Aware — every pattern considers Web Vitals.

Stack Compatibility

Technology	Version	Key Features
React	19+	Server Components, `use()`, Compiler
Next.js	15+	App Router, Server Actions, PPR
TypeScript	5+	Strict mode, no `any`
Tailwind CSS	v4	CSS-first config, `@theme`
Supabase	Latest	Auth, RLS, Edge Functions, Realtime
TanStack Query	v5	`queryOptions()`, prefetch, hydration
Zustand	v5	Slices, immer, selective persist
Zod	v3	Input validation, type inference
Framer Motion	v11	Animations, gestures, layout

Keeping Up to Date

cd ~/cursor-kenji && git pull && ./install.sh

Auto-sync via crontab

0 9 * * * cd ~/cursor-kenji && git pull origin main && ./install.sh --quiet

Contributing

See docs/CONTRIBUTING.md for how to add skills, commands, and rules.

See docs/CATALOG.md for the full reference with trigger phrases.

MIT License — Use freely, modify freely, share freely.

Built by @kensaurus. Enhanced continuously.

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
agents		agents
commands		commands
docs		docs
mcp		mcp
notepads		notepads
rules		rules
shell-aliases		shell-aliases
skills-cursor		skills-cursor
skills		skills
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
install.sh		install.sh

Folders and files

Latest commit

History

Repository files navigation

cursor-kenji

How It All Fits Together

What's New in 2026

May 2026 — Native RN Loop & Cross-Surface UI Latest

May 2026 — Housekeep & Direct-Tone Pass

Apr 2026 — The Enhance Family

Earlier 2026

What's Inside

Core

Configuration

Quick Start

What the installer does

How to Use This Toolkit

The four ways the agent gets help

Common daily flows

How to invoke a skill (without overthinking it)

How the new "Enhance" family works

Setting up a new project

Tips for getting the most out of skills

Skills (60)

Skill Categories at a Glance

Enhance New Apr-May 2026

Design & Frontend

Data & Creative

Backend & Database

Architecture & Quality

Engineering Practices New in 2026

Process & Documentation

Audits & Monitoring New

Debugging New

Testing & QA New

Product & Documentation New

File Handling New

Meta & Tooling

Commands (13)

Development Workflow

Coding Workflow

Research & Documentation

Bundles

Subagents (5)

MCP Servers (16)

Project Rules Starter Pack

Shell Helpers

Design Principles

Stack Compatibility

Keeping Up to Date

Contributing

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

May 2026 — Native RN Loop & Cross-Surface UI ^Latest

Enhance ^{New Apr-May 2026}

Engineering Practices ^{New in 2026}

Audits & Monitoring ^New

Debugging ^New

Testing & QA ^New

Product & Documentation ^New

File Handling ^New

Packages