Agent Skills

24 skills for shipping better software.

Quickstart

npx skills add mblode/agent-skills -g --agent codex claude-code -y

Works with Claude Code, OpenCode, Codex, and Cursor.

Why these skills exist

Code review catches logic bugs. Nobody checks the loading states, the type scale, or whether half the diff is AI slop. These skills do.

Skills

Architecture

• define-architecture: Folder structures, module contracts, and guardrails for TypeScript apps.
• scaffold-nextjs: Next.js turborepo with Blode UI, Ultracite, GitHub, and Vercel.
• scaffold-cli: TypeScript CLI package: ESM, tsdown, vitest, changesets, CI.
• multi-tenant-architecture: Tenant isolation, routing, and custom domains on Cloudflare or Vercel.

Design

• product-design: What the interface should do, before anyone builds it.
• ui-design: Visual direction, Tailwind builds, screenshot to markup, dark mode, responsive.
• ui-animation: Springs, gestures, easing, and curves pulled from a screen recording.
• presentation-creator: Dark-first decks with narrative arcs, speaker notes, and a QA pass.

Writing

• copywriting: Product and marketing copy, persuasion frameworks, AI-ism removal.
• docs-writing: Diataxis doc types, 52 rules, audit and writing modes.
• readme-creator: Writes or rewrites a README for the project type.

Personal voice training, evaluation, and long-form blog writing now live in the standalone ghostwriter project, where posts are drafted from the blog platform profile.

Quality

• ui-audit: React and Next.js UX audit: 95 rules, 12 playbooks, ship verdict.
• ax-audit: Agentic experience audit: 23 rules, ship verdict.
• dx-audit: Libraries, CLIs, and SDKs: root-cause findings, agent-friendly checks.
• typography-audit: 90 rules: punctuation, fonts, sizing, spacing, hierarchy, pairing.
• optimise-seo: Next.js SEO: sitemaps, metadata, structured data, hreflang, CWV.

Shipping

• planning: Builds the plan, then scores it to 5/5 before any code.
• pr-reviewer: Read-only diff review: bugs, structure, AI slop, security.
• tidy: The same sweep across four agents, but it applies the fixes.
• pr-creator: PRs with short human descriptions, Linear IDs, tidied commits.
• pr-babysitter: Watches an open PR: conflicts, CI, comments. Fixes what it can.
• autoship: npm releases with changesets: fix loop, CI watch, publish verify.

Authoring

• agents-md: Audits and refactors AGENTS.md and CLAUDE.md, with grades.
• agent-skills-creator: Creates and improves skills to the open format spec.

License

MIT

Crafted by Matthew Blode