Guides

Practical guides to Claude Code skills, Codex skills, the Agent Skills standard, CLAUDE.md, hooks, subagents and keeping a prompt library that works.

By the Kipt team · Updated 23 October 2026

Want to try it right away? The free SKILL.md generator turns any prompt into a skill, and the prompt library has copy-ready prompts for coding, writing and review.

Skills

14 Claude skills examples: copy-ready SKILL.md files for real work: 14 Claude skills examples with full SKILL.md files to copy: code review, commit messages, PR descriptions, release notes, tests, API docs, bug triage and more.
Agent Skills: the open SKILL.md standard, explained: Agent Skills explained: the SKILL.md format, progressive disclosure, which agents support the standard, and why one skill folder works in all of them.
Claude Code skills: what they are and how to install them: Claude Code skills explained: what a SKILL.md skill is, where skills live, how to install them from GitHub, run them with /name, and when to use one.
Claude Skills: the complete guide to what they are, where they work and how to make one: Claude Skills explained: what a skill is, where it works (Claude app, Claude Code, API), how to turn skills on, upload your own and write a SKILL.md.
Cursor skills: how Agent Skills work in Cursor: How Cursor skills work: the folders Cursor reads, how the agent picks a skill, how to run one, skills vs rules vs AGENTS.md, and sharing with Claude Code.
How to write a SKILL.md (with examples): How to create Claude skills (and Codex skills) with a SKILL.md: name and description rules, descriptions agents pick up, clear instructions and full examples.
Skill creator: how Anthropic's skill-creator works, plus other ways to build a skill: What Anthropic's skill creator does, how to install it in Claude Code and use it in Claude, and other ways to make a skill: by hand or with a free generator.
Superpowers skills: what's in obra/superpowers and how to install it: Superpowers skills explained: the open source obra/superpowers library, every skill it includes, the brainstorm, plan and TDD workflow, and install commands.
The best Claude Code skills: a curated list of collections worth installing: The best Claude Code skills, curated: Anthropic's official skills, frontend-design, Superpowers, Vercel and Trail of Bits collections, and how to install each.
The frontend-design skill: install it, use it, make your own: Anthropic's frontend design skill steers Claude Code away from generic AI-looking UI. What it says, how to install it, good prompts, and making your own.

Skill directory

Accessibility skill: audit and fix WCAG 2.2 issues with your coding agent: Addy Osmani's accessibility skill teaches coding agents to audit and fix web pages against WCAG 2.2. What it checks and how to install it.
Algorithmic art skill: generative p5.js art with Claude: Anthropic's algorithmic art skill has Claude write a generative art philosophy, then a seeded p5.js piece with sliders in one HTML file. How to install it.
Animation vocabulary skill: name any motion effect for your agent: The animation vocabulary skill turns a vague description of a UI motion effect into its exact name. What it does, how to install it, and how to use it.
API design principles skill: what it does and how to install it: This API design principles skill from wshobson/agents gives your agent REST and GraphQL design rules, a checklist and a FastAPI template. How to install it.
API design skill: what it does and how to install it: Addy Osmani's API design skill teaches your agent contract-first interfaces, one error format, boundary validation and idempotency keys. How to install it.
Apple design skill: Apple's fluid interface principles for web agents: This Apple design skill teaches your agent Apple's WWDC principles for springs, gestures, materials and type, for the web. What it does and how to install it.
Architecture patterns skill: Clean Architecture, hexagonal and DDD for your agent: The architecture patterns skill from wshobson/agents teaches your agent Clean Architecture, hexagonal ports and adapters, and DDD. How to install it.
Ask Matt skill: find the right Matt Pocock skill for your situation: The ask-matt skill is a router over mattpocock/skills: describe your situation and it names the next skill or flow to run. What it does and how to install it.
Better Auth skill: what it does and how to install it: The Better Auth skill (better-auth-best-practices) teaches your agent auth.ts setup, adapters, sessions, plugins and the auth CLI. What it does, how to install.
Brainstorming skill: what it does and how to install it: The brainstorming skill from Superpowers makes your coding agent ask what you want and why before it builds. What it does, its files, and how to install it.
Caveman skill: terse, answer-first replies from Claude Code and other agents: The caveman skill makes Claude Code, Codex and other agents answer first and drop filler, keeping code and errors exact. What it does and how to install it.
Claude API skill: Anthropic's reference skill for building with Claude: The claude api skill gives your agent current Claude API and SDK reference for seven languages plus cURL, and workflows like migrate. How to install and use it.
Cloudflare Agents SDK skill: build stateful agents with your coding agent: The Cloudflare Agents SDK skill gives your agent a docs map, core APIs and config gotchas for building agents with the agents package. Install and use it.
Cloudflare Workers skill: workers-best-practices for coding agents: The Cloudflare Workers skill makes your agent check current docs and flag 15 Workers anti-patterns, from floating promises to leaked secrets. How to install it.
Code quality skill: Addy Osmani's five-axis code review for agents: Addy Osmani's code-review-and-quality is a code quality skill that reviews each change on five axes with severity labels. What it does and how to install it.
Code review skill: Matt Pocock's two-axis review with parallel sub-agents: Matt Pocock's code review skill checks a diff against your repo's standards and the original spec, in two parallel sub-agents. What it does and how to install it.
Code simplification skill: Addy Osmani's behavior-preserving cleanup for agents: The code simplification skill from addyosmani/agent-skills makes your agent cut complexity without changing behavior, one tested step at a time. How to install.
Codebase design skill: what it does and how to install it: The codebase design skill gives your agent one vocabulary for deep modules: module, interface, seam, adapter, depth. What it does and how to install it.
Convex agent skills: the main convex skill that routes to the rest: The convex skill is the entry point to Convex agent skills: it routes your agent to the right convex-* skill. What it does, how to install it and its audits.
Convex create component skill: build reusable Convex components: The convex create component skill is Convex's official Agent Skill for building reusable components with their own tables. What it does and how to install it.
Convex expert skill: the rules agents follow for code in convex/: The convex expert skill is Convex's official Agent Skill for any code under convex/: function syntax, validators, indexes and imports. What it does and install.
Convex skill: convex-quickstart for Claude Code and Codex: convex-quickstart is Convex's official Agent Skill for starting a Next.js and Convex app from a one-line idea. What the Convex skill does and how to install it.
Copywriting skill for Claude Code: what it does and how to install it: Corey Haines' copywriting skill helps Claude Code write landing page and homepage copy without AI tells or made-up claims. What it does and how to install it.
Core Web Vitals skill: measure and fix LCP, INP and CLS with your agent: The core web vitals skill from addyosmani/web-quality-skills has your agent check field data, trace the page and fix LCP, INP and CLS. How to install and use.
Debugging and error recovery skill: Addy Osmani's stop-the-line triage for agents: The debugging and error recovery skill from addyosmani/agent-skills makes your agent reproduce, localize, fix and guard each bug. What it does, how to install.
Deploy to Vercel skill: what it does and how to install it: The deploy to Vercel skill lets Claude Code, Codex and claude.ai ship your app as a preview by git push, the Vercel CLI or a no-login script. How to set it up.
Deprecation and migration skill for Claude Code: what it does and how to install it: From Addy Osmani, a deprecation and migration skill Claude Code can load: strangler and adapter patterns, expand/contract schema changes, and how to install it.
Diagnosing bugs skill for Claude Code: what it does and how to install it: Matt Pocock's diagnosing-bugs is a debugging skill Claude Code follows: no guessing until a failing feedback loop exists. What it does and how to install it.
diagnosing-superpowers skill: find out why an agent session went wrong: The diagnosing-superpowers skill reads an agent session transcript and reports what went wrong, with line citations. What it does, its rules, how to install it.
Differential review skill from Trail of Bits: what it does and how to install it: Trail of Bits' differential review skill gives Claude Code a security review of PRs and diffs: git blame, blast radius, test gaps and a report file.
Dispatching parallel agents skill: run independent fixes at once: The dispatching parallel agents skill from Superpowers splits independent problems across subagents that work at the same time. How it works and how to install.
Docker skill for Claude Code: the multi-stage Dockerfile skill and how to install it: GitHub's multi-stage-dockerfile Docker skill helps Claude Code write smaller, safer Dockerfiles. What it covers and misses, and how to install and extend it.
Documentation and ADRs skill: Addy Osmani's skill for recording the why: The documentation and ADRs skill from addyosmani/agent-skills has your agent write decision records, why-comments and API docs. What it does, how to install.
Domain modeling skill: what it does and how to install it: The domain modeling skill from mattpocock/skills keeps a GLOSSARY.md and ADRs current as you design with an agent. What it does, how to install and use it.
Durable Objects skill: Cloudflare's skill for stateful Workers: Cloudflare's Durable Objects skill teaches your agent the rules for stateful objects: one per entity, SQLite storage, RPC, alarms and tests. Install and use it.
Emil Kowalski skill (emil-design-eng): UI polish and animation rules for agents: The Emil Kowalski skill, emil-design-eng, gives your agent his rules for animation, easing, springs and UI polish. What it does and how to install it.
Excalidraw skill: generate diagrams from text with your coding agent: The Excalidraw skill from GitHub's awesome-copilot turns plain-language requests into editable .excalidraw diagrams. What it does and how to install it.
Executing plans skill: what it does and how to install it: The executing plans skill from Superpowers has your agent run a written plan itself, test-first, with one final review. What it does and how to install it.
Expo native UI skill: what it does and how to install it: Expo's official expo-native-ui skill teaches your coding agent native-feeling screens: semantic colors, SF Symbols, controls and safe areas. Install and use it.
Expo skill: expo-dev-client and the official Expo skills for agents: The most installed Expo skill, expo-dev-client, helps your coding agent build Expo dev clients locally or via TestFlight. What it does and how to install it.
Fastify skill for Claude Code: fastify-best-practices: The Fastify skill, fastify-best-practices, is Matteo Collina's Agent Skill for Fastify APIs: plugins, schemas, hooks, testing. What it does and how to install.
Find animation opportunities skill: where your UI should move, and where it shouldn't: The find animation opportunities skill sweeps your UI for places that should animate, rejects the rest, and gives exact values. What it does, how to install.
Find bugs skill: Sentry's branch review for Claude Code and Codex: Sentry's find-bugs skill reviews your branch diff for security holes, bugs and quality issues, then reports without editing. What it checks and how to install.
Finishing a development branch skill: merge, PR or keep, safely: The finishing a development branch skill from Superpowers verifies tests, then lets you merge, open a PR or keep the branch, and cleans up worktrees safely.
Firebase Auth skill: firebase-auth-basics explained: firebase-auth-basics is Firebase's official Auth skill for agents: sign-in providers, firebase.json auth config, client SDKs and rules. What it does and install.
Firebase skill for Claude Code: firebase-basics explained: firebase-basics is Firebase's official Agent Skill for CLI login, project setup and app config files. What the Firebase skill does and how to install it.
Frontend UI engineering skill for Claude Code: what it does and how to install it: Addy Osmani's frontend UI engineering skill steers Claude Code toward accessible, responsive UI that avoids the AI look. What it covers and how to install it.
Git commit skill for Claude Code: Conventional Commits from your diff: GitHub's git commit skill lets Claude Code, Copilot and Codex read your diff, stage related files and write a Conventional Commits message. How to install it.
Git worktrees skill: how Superpowers isolates agent work on a new branch: The git worktrees skill from Superpowers gives a coding agent an isolated workspace before it builds. How it picks a worktree, sets up, and how to install it.
GitHub Actions skill: workflow templates for your coding agent: This GitHub Actions skill from wshobson/agents gives your agent CI/CD workflow templates for tests, Docker, Kubernetes and approvals. How to install and use it.
Grill me skill: what it does and how to install it: The grill me skill from mattpocock/skills makes your agent interview you about a plan until every decision is settled. What it does and how to install it.
Grill with docs skill: what it does and how to install it: The grill with docs skill interviews you about a plan and writes GLOSSARY.md terms and ADRs into your repo as you go. What it does and how to install it.
Grilling skill: what it does and how to install it: The grilling skill from mattpocock/skills interviews you in rounds of numbered questions until a plan has no open decisions. How it works and how to install it.
GSAP performance skill: official rules for smooth GSAP animation: The GSAP performance skill teaches agents to animate transforms, use quickTo, batch DOM work and clean up ScrollTriggers to avoid jank. What it covers and how to install.
GSAP ScrollTrigger skill: official scroll animation skill for agents: The GSAP ScrollTrigger skill teaches coding agents scroll-linked animation, pinning, scrub, batch and refresh rules. What it covers and how to install it.
GSAP skill (gsap-core): the official GSAP animation skill for agents: gsap-core is GreenSock's official GSAP skill: tweens, easing, stagger and matchMedia rules for Claude Code, Codex and Cursor. What it does and how to install.
GSAP timeline skill: the official sequencing skill for coding agents: The GSAP timeline skill teaches agents gsap.timeline(), the position parameter, labels, nesting and playback control. What it covers and how to install it.
Handoff skill: what it does and how to install it: The handoff skill from mattpocock/skills compacts a conversation into a file another agent can pick up. What it does, when to use it and how to install it.
Hugging Face datasets skill: explore Hub datasets from your agent: The Hugging Face datasets skill teaches your agent the Dataset Viewer API: splits, rows, search, filters, parquet and stats. What it does and how to install.
Hugging Face skill: what hf-cli does and how to install it: The top Hugging Face skill is hf-cli: it teaches Claude Code, Codex and Cursor every hf command for models, datasets, Spaces and jobs. How to install it.
Humanizer skill: what it does and how to install it: The humanizer skill rewrites AI-sounding text so it reads like a person wrote it, using 26 patterns from Wikipedia's Signs of AI writing. How to install it.
HyperFrames CLI skill: teach your agent the hyperframes command line: The HyperFrames CLI skill teaches your agent the hyperframes commands by job: scaffold, check, snapshot, preview, render. What it does and how to install it.
HyperFrames figma skill: turn Figma designs into video with your agent: The HyperFrames figma skill has your agent import Figma assets, brand tokens, components and storyboards into an HTML video. What it does and how to install it.
HyperFrames registry skill: ready-made video blocks and components for your agent: The HyperFrames registry skill has your agent install about 450 ready-made video blocks and components as editable source. What it does and how to install it.
HyperFrames skill: make videos from HTML with Claude Code and other agents: The HyperFrames skill teaches your agent to build, check and render videos as HTML with HeyGen's open-source framework. What it does and how to install it.
Impeccable skill: what it does and how to install it: The Impeccable skill gives your agent one /impeccable command with 24 design commands and an anti-pattern detector. What it does, what runs, how to install.
Implement skill: build a ticket or spec with tdd and code review: The implement skill from mattpocock/skills builds one ticket or spec: tdd at agreed seams, typechecks, code review, then a commit. How to install and use it.
Implement spec skill: what it does and how to install it: The implement spec skill lands a whole spec in one run, sending tickets to parallel subagents in git worktrees. What it does, how to install and use it.
Improve animations skill: audit your app's motion and plan the fixes: The improve animations skill audits every animation in a codebase and writes self-contained fix plans any agent can run. What it does and how to install it.
Improve codebase architecture skill: what it does and how to install it: Matt Pocock's improve codebase architecture skill scans your repo for shallow modules and writes an HTML report of refactor candidates. How to install it.
Incremental implementation skill: Addy Osmani's thin-slice build loop for agents: The incremental implementation skill from addyosmani/agent-skills has your agent build in thin slices: implement, test, verify, commit. How to install it.
Internal comms skill: Anthropic's skill for status updates and newsletters: Anthropic's internal comms skill gives Claude fixed formats for 3P updates, company newsletters and FAQ roundups. What's inside, how to install and use it.
Kubernetes skill: the k8s manifest generator and how to install it: This Kubernetes skill from wshobson/agents helps your agent write Deployments, Services, ConfigMaps and Secrets for production. What it covers, how to install.
Matt Pocock skills setup: what /setup-matt-pocock-skills does: Matt Pocock skills setup explained: what /setup-matt-pocock-skills asks, which docs/agents files it writes, and how to install and run it once per repo.
media-use skill: music, voiceover, images and grades for HyperFrames videos: The media-use skill has your agent find or make music, voiceover, images, logos, avatars and grades for HyperFrames videos, with HeyGen or free local tools.
Modern Python skill: uv, ruff and ty setup for coding agents: Trail of Bits' modern Python skill makes your agent set up Python with uv, ruff, ty, pytest and prek, and migrate off pip or Poetry. How to install it.
Neon Postgres skill: what it does and how to install it: The Neon Postgres skill teaches your agent pooled vs direct connections, branch-tested migrations, neon inspect db checks and Lakebase Search. How to install.
Neon skill: the official overview skill for building on Neon: The official neon skill gives your coding agent Neon's setup path, neon.ts config, branch-first workflow and backend primitives. Install and use it.
Next.js bundle optimizer skill: shrink client bundles with your agent: next-bundle-optimizer is the official Next.js skill that uses next analyze to find oversized client bundles and fix them one verified change at a time.
Next.js cache components optimizer skill: make routes instant: next-cache-components-optimizer is the official Next.js skill that grows a route's static shell, proven by a failing instant() test that it turns green.
Next.js skill for Claude Code: next-dev-loop and how to install it: next-dev-loop is the official Next.js skill that makes Claude Code verify each edit in a running app through /_next/mcp and agent-browser. How to install it.
Performance optimization skill for Claude Code: what it does and how to install it: Addy Osmani's performance optimization skill makes your agent measure, fix one bottleneck, re-measure and revert neutral changes. What it does, how to install.
Performance skill: measure first, then speed up your site with an agent: The performance skill from addyosmani/web-quality-skills has your agent measure field and lab data, trace the page, then fix loading and runtime issues.
Pick UI library skill: Emil Kowalski's curated frontend library picks: The pick UI library skill has your agent choose a frontend library from Emil Kowalski's curated list, not hand-roll one. What it covers and how to install.
Planning and task breakdown skill for Claude Code: what it does and how to install it: Addy Osmani's planning and task breakdown skill turns a spec into small, ordered tasks with acceptance criteria and checkpoints. What it does, how to install.
Playwright best practices skill: a testing reference for your coding agent: The Playwright best practices skill from Currents.dev gives coding agents 58 reference guides for writing, debugging and fixing Playwright tests. How to install it.
Playwright CLI skill: Microsoft's official Playwright skill for coding agents: Microsoft's official Playwright skill lets Claude Code and other agents drive a browser with playwright-cli and write Playwright tests. How to install it.
Postgres best practices skill: what it does and how to install it: Supabase's postgres best practices skill gives your agent 31 rules for indexes, RLS, pooling and locking on any Postgres. What it does and how to install it.
Prisma skill: the prisma-client-api skill and how to install it: The most-installed Prisma skill, prisma-client-api, gives your agent Prisma Client queries, filters, relations, transactions and safe raw SQL. How to install.
Property-based testing skill: Trail of Bits' skill for coding agents: Trail of Bits' property-based testing skill helps your agent write, review and triage Hypothesis, fast-check and proptest suites. What it does and how to install.
Prototype skill: what it does and how to install it: The prototype skill builds throwaway code to answer one design question: a clickable HTML logic demo or several UI variants. What it does and how to install it.
React best practices skill: what it does and how to install it: Vercel's React best practices skill gives your agent 70 React and Next.js performance rules, ranked by impact. What it does and how to install it.
React composition patterns skill: what it does and how to install it: Vercel's React composition patterns skill teaches your agent compound components and lifted state instead of boolean props. What it does and how to install it.
React Native skill: Vercel's React Native and Expo rules for coding agents: Vercel's React Native skill gives your agent 36 React Native and Expo rules for lists, animation, navigation and native UI. What it does and how to install it.
React view transitions skill: what it does and how to install it: Vercel's React view transitions skill teaches your agent the ViewTransition component, transition types and Next.js patterns. What it does and how to install.
Receiving code review skill: how Superpowers makes agents check feedback first: The receiving code review skill from Superpowers makes a coding agent verify review feedback against the code before acting, and push back when it's wrong.
Remotion skill: make videos in React with Claude Code and other agents: The Remotion skill (remotion-best-practices) teaches your agent to create, preview and render videos in React with Remotion. What it does and how to install it.
Remotion to HyperFrames skill: port existing video code with your agent: The remotion-to-hyperframes skill has your agent port Remotion and other video code into HyperFrames and check it frame by frame. What it does and how to install it.
Requesting code review skill: how Superpowers gets agents a second look: The requesting code review skill from Superpowers has an agent dispatch a reviewer subagent with a git range and the plan. What it does and how to install it.
Review animations skill: Emil Kowalski's strict motion review for agents: This review animations skill makes your agent review animation code against Emil Kowalski's rules and end with Block or Approve. What it does, how to install.
Security and hardening skill for Claude Code: what it does and how to install it: From Addy Osmani, a security skill Claude Code can load: threat modeling, OWASP Top Ten defenses and a review checklist. What it covers and how to install it.
Semgrep skill from Trail of Bits: what it does and how to install it (plus CodeQL): Trail of Bits' semgrep skill and codeql skill run security scans from Claude Code with an approval gate and SARIF output. What they do and how to install them.
Sentry skill: code-simplifier and the getsentry/skills collection: The most-installed Sentry skill that passes every skills.sh audit is code-simplifier. What it does, how to install Sentry's skills and which audits fail.
SEO AEO GEO skill: optimize for AI search with Claude Code: The seo aeo geo skill from RampStack gives Claude Code a 5-layer plan for AI overviews, LLM citations and llms.txt. What it does and how to install it.
SEO skill for Claude: Addy Osmani's evidence-led SEO audit: The SEO skill Claude Code, Codex and Cursor can load from addyosmani/web-quality-skills: Lighthouse-led crawl, metadata and schema checks. How to install it.
shadcn skill: what it does and how to install it: The official shadcn skill gives your agent your project's shadcn/ui config and the rules for adding and composing components. What it does, how to install.
Slack GIF creator skill: animated emoji and GIFs with Claude: Anthropic's slack gif creator skill gives Claude Python helpers and Slack size rules for animated emoji and message GIFs. What's inside and how to install it.
Spec-driven development skill for Claude Code: what it does and how to install it: Addy Osmani's spec-driven development skill makes your agent write and get approval for a spec before any code. What the spec covers and how to install it.
Stripe skill: what it does and how to install it: Stripe's official Stripe skill, stripe-best-practices, steers your agent to Checkout Sessions, webhooks and restricted API keys. What it does, how to install.
Subagent driven development skill: what it does and how to install it: Subagent driven development from Superpowers runs your plan with a fresh subagent per task and a review after each one. How it works and how to install it.
Supabase skill: the official agent skill for every Supabase task: The official Supabase skill gives your coding agent a security checklist, CLI and MCP gotchas, and a schema change workflow for Supabase. Install and use it.
SwiftUI skill for Claude Code: swiftui-expert-skill explained: The SwiftUI skill by Antoine van der Lee helps your agent write and review SwiftUI: state, lists, performance, Liquid Glass and traces. How to install it.
Systematic debugging skill: what it does and how to install it: The systematic debugging skill from Superpowers makes Claude Code find the root cause before it fixes a bug. The four phases, the files, and install commands.
TDD skill for Claude Code: what it does and how to install it: From Matt Pocock, a tdd skill Claude Code can load: a red-green testing loop at agreed seams, with rules for good tests and mocks. How to install and use it.
Teach skill: learn any topic with Claude Code across many sessions: Matt Pocock's teach skill turns a folder into a learning workspace: a mission, vetted sources, short HTML lessons, records. What it does and how to install it.
Terraform module skill: HashiCorp's refactor-module for coding agents: HashiCorp's Terraform module skill, refactor-module, splits a monolithic config into reusable modules with moved blocks. What it does and how to install.
Terraform skill from HashiCorp: what it does and how to install it: HashiCorp's official Terraform skill makes Claude Code write HCL that follows the HashiCorp style guide. What it covers, terraform-test, and how to install it.
Test driven development skill: how Superpowers enforces TDD for agents: The test driven development skill from Superpowers makes coding agents write a failing test before any production code. What it does and how to install it.
The agent-browser skill: browser automation for Claude Code, Codex and Cursor: The agent browser skill from Vercel Labs lets coding agents open sites, click, fill forms and take screenshots with a CLI. What it does and how to install it.
The brand-guidelines skill: a model for your own brand skill: Anthropic's brand guidelines skill applies its brand colors and fonts to artifacts. What's inside, how to install it, and how to adapt it for your own brand.
The canvas-design skill: posters and art as PDF or PNG: Anthropic's canvas design skill has your agent write a design philosophy, then render it as a poster or art piece in PDF or PNG. What's inside, how to install.
The Claude docx skill: Word documents, explained: The Claude docx skill creates, edits and reviews Word documents, with tracked changes and comments. What's in it, how to install it, and its license.
The Claude Excel skill (xlsx): what it does and how to install it: The Claude Excel skill builds and edits .xlsx files with live formulas and checks them for errors. What's in it, how to install it, and its license limits.
The Claude PDF skill: what it does and how to install it: The Claude PDF skill reads, merges, splits, fills and creates PDFs. What is in Anthropic's pdf folder, how to install it, and its license limits.
The Claude PowerPoint skill (pptx): what it does and how to install it: The Claude PowerPoint skill builds and edits .pptx decks, fills templates and checks every slide visually. What's in it, how to install it, and its license.
The design-taste-frontend skill (Taste Skill): art direction for coding agents: The taste skill (design-taste-frontend) gives coding agents art direction and a real-browser check against generic AI pages. What it does, how to install it.
The doc-coauthoring skill: write specs and proposals with Claude: Anthropic's doc coauthoring skill walks you and Claude through context gathering, section-by-section drafting and reader testing for specs, PRDs and proposals.
The find-skills skill: let your agent find and install skills: The find skills skill from Vercel teaches your agent to search skills.sh, vet results and install skills with the skills CLI. What it does and how to install.
The HyperFrames slideshow skill: build live presentation decks with your coding agent: The HyperFrames slideshow skill has your agent build a live HTML deck with reveals, branches and presenter notes. What it does and how to install it.
The mcp-builder skill: build MCP servers with Claude: Anthropic's mcp builder skill guides your agent through planning, coding, testing and evaluating an MCP server in TypeScript or Python. How to install it.
The reddit-automation skill: find Reddit threads and draft honest, disclosed replies: The reddit automation skill finds threads you can genuinely help with and drafts disclosed replies you post yourself. How to install it, and Reddit rules.
The theme-factory skill: 10 ready themes for slides and docs: Anthropic's theme factory skill gives your agent 10 color and font themes for slides, docs and HTML pages, or makes a new one. What's inside and how to install.
The web-artifacts-builder skill: complex React artifacts in one HTML file: Anthropic's web artifacts builder skill scaffolds a React, Tailwind and shadcn/ui project, then bundles it into one HTML file. What's inside, how to install.
The webapp-testing skill: test local web apps with Playwright: Anthropic's webapp testing skill lets your agent drive a local web app with Python Playwright: screenshots, logs, clicks. What's inside and how to install it.
tmux skill for Claude Code: drive vim, REPLs and git rebase -i: The using-tmux-for-interactive-commands skill lets your agent run vim, REPLs and git rebase -i in detached tmux sessions. How it works and how to install it.
To-spec skill: turn an agent conversation into a spec: The to-spec skill from mattpocock/skills turns your conversation into a spec on the issue tracker, with no new interview. What it writes and how to install it.
Triage skill: what it does and how to install it: The triage skill from mattpocock/skills moves incoming issues and PRs through labelled states and writes agent-ready briefs. What it does and how to install it.
UI UX Pro Max skill: what it does and how to install it: The UI UX Pro Max skill gives your coding agent a local, searchable database of UI styles, palettes, fonts and UX rules. What it does and how to install it.
Upgrade Stripe skill: what it does and how to install it: Stripe's official upgrade-stripe skill guides your agent through API version, webhook, SDK and Stripe.js upgrades. What it does and how to install it.
using-superpowers skill: the bootstrap that makes Superpowers fire: The using-superpowers skill is the Superpowers bootstrap that makes your agent check for skills before any reply. How it loads, what it says, how to install it.
Vercel CLI skill (vercel-cli-with-tokens): what it does and how to install it: The Vercel CLI skill, vercel-cli-with-tokens, lets your agent deploy and manage Vercel projects with a VERCEL_TOKEN instead of vercel login. How to install it.
Vercel optimize skill: what it does and how to install it: The Vercel optimize skill audits a deployed Vercel project's cost and performance from real metrics, then ranks verified fixes. What it does and how to install.
Verification before completion skill: make agents prove work is done: The verification before completion skill from Superpowers makes a coding agent run the check and read the output before it says done. Rules, install and use.
Wayfinder skill: plan work too big for one agent session: The wayfinder skill from mattpocock/skills plans work too big for one agent session as a map of decision tickets on your tracker. How to install and use it.
Web best practices skill: security, compatibility and code quality checks: The web best practices skill from addyosmani/web-quality-skills runs Lighthouse Best Practices checks and fixes HTTPS, CSP, deprecated APIs and console errors.
Web design guidelines skill: what it does and how to install it: Vercel's web design guidelines skill audits UI code against its Web Interface Guidelines and reports issues as file:line. What it does and how to install it.
Web performance skill: Cloudflare's Core Web Vitals audit for coding agents: Cloudflare's web performance skill (web-perf) has your agent trace a page, read Core Web Vitals and Lighthouse metrics, and rank fixes. How to install it.
Web quality audit skill: a Lighthouse-style review run by your agent: The web quality audit skill from addyosmani/web-quality-skills audits performance, accessibility, SEO, best practices and agentic browsing with evidence.
Wrangler skill: Cloudflare's CLI skill for coding agents: Cloudflare's wrangler skill makes your agent check the project's Wrangler version and current docs before it deploys, edits config or sets secrets. Install it.
Writing clearly and concisely skill: Strunk's rules for your coding agent: The writing-clearly-and-concisely skill gives your agent Strunk's Elements of Style for docs, commits and error text. What it does and how to install it.
Writing guidelines skill: review docs against Vercel's writing handbook: Vercel's writing guidelines skill has your agent review docs and prose against the Vercel writing handbook, with file:line findings. How to install and use it.
Writing plans skill: what it does and how to install it: The writing plans skill from Superpowers turns a spec into a task-by-task implementation plan for your agent. What a plan contains and how to install it.
Writing skills skill: how Superpowers tests a SKILL.md before you ship it: The writing skills skill from Superpowers treats writing a SKILL.md as test-driven development: watch an agent fail first, then write it. Install and use.

Claude Code

Best MCP servers: a curated list by job, with Claude Code install commands: The best MCP servers grouped by job: GitHub, Playwright, Context7, Supabase, Sentry, Figma, Linear, Notion, with Claude Code commands and security notes.
Claude Code agent teams: how to run a team of Claude sessions: Claude Code agent teams explained: enable the experimental flag, spawn teammates, share a task list, pick display modes and models, and know when subagents fit.
Claude Code auto mode: how the classifier works and how to control it: Claude Code auto mode explained: how the safety classifier decides, what it blocks, how to turn it on or off, autoMode settings, supported models and limits.
Claude Code best practices: 12 habits that actually help: Practical Claude Code best practices and tips: plan first, keep CLAUDE.md short, manage context, give Claude tests to run, set permissions, review the output.
Claude Code cheat sheet: commands, flags, shortcuts and files: A one-page Claude Code cheat sheet: CLI commands and flags, slash commands, keyboard shortcuts, permission modes, memory and settings files, MCP and plugins.
Claude Code checkpoints: how /rewind works and what it can undo: Claude Code checkpoints explained: how /rewind and Esc Esc restore code and conversation, the summarize options, what is not tracked, and how to turn it off.
Claude Code computer use: let Claude click, type and see your screen: Claude Code computer use lets Claude open apps, click and screenshot your screen. How to enable it in the CLI or desktop app, approve apps and stay safe.
Claude Code context window: what fills it and how to manage it: How the Claude Code context window works: what loads before you type, /context, /compact, auto-compaction, what survives it, 1M token models, and tips.
Claude Code cost: how to track and cut what you spend: Claude Code cost explained: read /usage, find what burns tokens, and spend less with the right model, lower effort, lean context and a few cheap habits.
Claude Code desktop app: the Code tab, explained: How the Claude Code desktop app works: install it, open the Code tab, run parallel sessions, review diffs, preview your app and move sessions over from the CLI.
Claude Code enterprise: managed settings, policies and admin setup: Claude Code enterprise setup: pick a provider, deliver managed settings, enforce permissions, MCP and model policy, track usage, and verify it with /status.
Claude Code fast mode: faster Opus responses with /fast: Claude Code fast mode makes Opus up to 2.5x faster at a higher price. How to turn on /fast, what it costs, who can use it and when it is worth paying for.
Claude Code GitHub Actions: setup, workflows and @claude mentions: Claude Code GitHub Actions explained: run /install-github-app or set up by hand, copy the workflow YAML, add secrets, use @claude mentions and keep costs down.
Claude Code GitLab CI/CD: run Claude in your pipelines and MRs: Set up Claude Code with GitLab CI/CD: add a job to .gitlab-ci.yml, store your API key, trigger it from merge requests or @claude comments, and use Bedrock or Vertex.
Claude Code headless mode: run claude -p in scripts and CI: Claude Code headless mode explained: run claude -p in scripts and CI, get json or stream-json output, pre-approve tools, use bare mode and continue sessions.
Claude Code hooks: events, config and copy-ready examples: Claude Code hooks run your commands at fixed points, like before a tool call. All 33 events, the five hook types, exit codes and six copy-ready examples.
Claude Code in a devcontainer: setup, persistence and security: Claude Code devcontainer guide: add the official Dev Container Feature, keep login across rebuilds, enforce managed settings and restrict network egress.
Claude Code in JetBrains IDEs: install the plugin, features and fixes: Claude Code JetBrains plugin guide: install it in IntelliJ, PyCharm or WebStorm, view diffs in the IDE, share selections, connect with /ide and fix WSL issues.
Claude Code in VS Code: install the extension, features and shortcuts: Claude Code VS Code extension guide: install it in VS Code or Cursor, review diffs, @-mention files, key shortcuts and settings, and when to use the CLI.
Claude Code keyboard shortcuts: the full reference: All Claude Code keyboard shortcuts in one place: general controls, text editing, multiline input, vim mode, Option as Meta on macOS and custom keybindings.
Claude Code MCP: add, manage and secure MCP servers: Claude Code MCP guide: the claude mcp add command, local, project and user scopes, HTTP and stdio transports, useful servers, security, and MCP vs skills.
Claude Code memory: how CLAUDE.md and auto memory work: How Claude Code memory works: CLAUDE.md locations and load order, @ imports, .claude/rules, auto memory, the /memory command, and what survives /compact.
Claude Code models: aliases, /model, opusplan and effort: Claude Code models explained: switch with /model, use aliases like opus, sonnet and fable, try opusplan, set effort, and see the default model for each plan.
Claude Code not working? Fixes for install, login, MCP and more: Claude Code not working? Documented fixes for command not found, login and 403 errors, /doctor, permission surprises, MCP servers that won't load, and slowness.
Claude Code on Amazon Bedrock, Google Cloud's Agent Platform and Microsoft Foundry: Claude Code Bedrock guide, plus Google Cloud's Agent Platform (Vertex AI) and Microsoft Foundry: setup wizard, env vars, IAM, model pinning and limits.
Claude Code on Windows: native install, WSL and fixes: Claude Code Windows guide: install natively from PowerShell, CMD or WinGet, or inside WSL. Git Bash requirements, PATH fixes and common Windows errors solved.
Claude Code OpenTelemetry: monitor usage, costs and tool activity: Set up Claude Code OpenTelemetry: enable telemetry, choose exporters, enforce it with managed settings, and use the metrics, events and traces it exports.
Claude Code output styles: built-in styles, switching, and custom styles: Claude Code output styles explained: the four built-in styles, how to switch with /output-style, how to write a custom style, and how it differs from CLAUDE.md.
Claude Code permissions: allow, ask and deny rules, modes and safe defaults: Claude Code permissions explained: allow, ask and deny rule syntax, permission modes, the /permissions dialog, managed policies and a safe default setup.
Claude Code plan mode: how to make Claude plan before it edits: How Claude Code plan mode works: turn it on with Shift+Tab or a flag, what Claude can do while planning, how to approve or edit a plan, and when it is worth it.
Claude Code plugins: what they are, how to install them, how to make one: Claude Code plugins bundle skills, subagents, hooks and MCP servers into one install. How marketplaces work, how to build one, and plugins vs skills.
Claude Code pricing: plans, usage limits and API costs: Claude Code pricing explained: which Claude plans include it, Pro vs Max vs Team seats, API pay-as-you-go, usage limits, checking usage and spending less.
Claude Code prompt caching: what breaks the cache and how to check it: How Claude Code prompt caching works, which actions invalidate the cache, the 5-minute and 1-hour TTLs, how to check your cache hit rate, and how to change it.
Claude Code proxy and LLM gateway setup: base URL, tokens and HTTPS_PROXY: Claude Code proxy setup: point Claude Code at an LLM gateway with ANTHROPIC_BASE_URL, pick the right token variable, set HTTPS_PROXY and fix gateway errors.
Claude Code Remote Control: continue a local session from your phone: Claude Code Remote Control lets you drive a session running on your computer from claude.ai/code or the Claude mobile app. Setup, commands, security and limits.
Claude Code review: /code-review, /security-review and PR reviews: Claude Code review options compared: the bundled /code-review (alias /review), /security-review, ultrareview, managed GitHub Code Review and review workflows.
Claude Code sandbox: /sandbox, filesystem and network isolation: How the Claude Code sandbox works: turn it on with /sandbox, what filesystem and network isolation block, supported platforms, key settings and its limits.
Claude Code scheduled tasks: /loop, Desktop tasks and cloud schedules: Claude Code scheduled tasks explained: run prompts on repeat with /loop, schedule local Desktop tasks, or set up cloud scheduled tasks (formerly routines).
Claude Code security: safeguards, prompt injection and security reviews: How Claude Code security works: permission modes, sandboxing, prompt injection safeguards, MCP trust, /security-review, security plugins and best practices.
Claude Code sessions: how to resume, name, branch and export conversations: How to resume a Claude Code session with --continue, --resume and /resume, name sessions, branch a conversation, export transcripts and find where they are stored.
Claude Code settings: settings.json files, precedence and useful keys: Claude Code settings explained: where each settings.json file lives, which one wins, the most useful keys, environment variables, and how /config works.
Claude Code setup: install, sign in and your first session: Claude Code setup for beginners: install on macOS, Windows or Linux, sign in, first prompts, CLAUDE.md with /init, permissions and your first skill.
Claude Code slash commands: built-ins and custom commands: The Claude Code slash commands worth knowing, and how custom commands work now that they have merged into skills: files, arguments, frontmatter and examples.
Claude Code statusline: set up a custom status line: Set up a Claude Code statusline with /statusline or settings.json, learn the JSON fields your script gets, and copy bash scripts for git, context and cost.
Claude Code subagents: how to create and use them: Claude Code subagents run tasks in their own context with their own tools. How to create one, every frontmatter field, four examples, and subagents vs skills.
Claude Code terminal setup: Shift+Enter, notifications, themes and more: Claude Code terminal setup guide: fix Shift+Enter, enable Option as Meta, get notifications, configure tmux, pick a theme, add a status line and go fullscreen.
Claude Code tips: 20 practical habits that save time and tokens: 20 Claude Code tips checked against the official docs: shortcuts, CLAUDE.md, plan mode, /compact and /clear, subagents, skills, hooks and headless runs.
Claude Code tools: the built-in tool list and how to control it: Claude Code tools explained: the built-in tools like Bash, Read, Edit and WebFetch, which ones ask permission, how to allow or remove them, and key limits.
Claude Code voice mode: how to dictate prompts with /voice: Claude Code voice mode lets you speak prompts in the CLI. How to turn on /voice, hold vs tap mode, change the key and language, and fix microphone errors.
Claude Code with Chrome: set up browser automation from the terminal: Claude Code Chrome guide: connect the Claude in Chrome extension with claude --chrome, test local web apps, read console errors, set permissions and fix issues.
Claude Code worktrees: run parallel sessions with git worktrees: How Claude Code worktrees work: start parallel sessions with claude --worktree, copy .env files, clean up, resume, isolate subagents and tune worktree settings.
CLAUDE.md: what to put in it, where it goes, and a template: A Claude md file (CLAUDE.md) gives Claude Code standing instructions each session: where it lives, what to put in it, a template and how it compares to skills.
How to create a Claude Code plugin marketplace and host it: How to create Claude Code plugin marketplaces: write marketplace.json, pick plugin sources, validate, host on GitHub or a URL, and ship updates safely.
Ultrathink in Claude Code: what it does now, and effort levels explained: What ultrathink does in Claude Code today, how it differs from effort levels and extended thinking, how to toggle thinking, and when deep reasoning pays off.

Claude

Claude Agent SDK: install it, build an agent, and control tools, permissions, MCP, skills and sessions: Claude Agent SDK guide: install the Python or TypeScript package, run your first agent, and set up tools, permissions, MCP servers, skills and sessions.
Claude Artifacts: what they are, how to make, share and remix them: Claude Artifacts explained: what counts as an artifact, templates, editing and export, sharing and publishing, AI-powered artifacts, storage and plans.
Claude Code on the web: run cloud sessions at claude.ai/code and move them to your terminal: Claude Code web sessions run coding tasks in cloud VMs at claude.ai/code. How it works, GitHub repos, environments, plans and moving sessions to your terminal.
Claude desktop app: download, system requirements, features and plans: The Claude desktop app for Mac, Windows and Linux: download and install steps, extensions, connectors, Claude Code, Cowork, and which plans get what.
Claude Projects: what they are, how project knowledge works, and how to write good instructions: Claude Projects explained: which plans have them, project knowledge and file limits, how RAG search works, team sharing and instructions that work.

Codex

AGENTS.md: what it is, which agents read it, and a template: AGENTS.md is an open Markdown file that tells coding agents how to work in your repo. Which agents read it, where it goes, nesting, and a template.
Codex CLI: install, commands, config and tips: Codex CLI guide: install with npm or Homebrew, sign in, sandbox and approval modes, key flags, slash commands, config.toml, MCP, skills and codex exec.
Codex Cloud: environments, tasks, diff reviews and plans: Codex Cloud guide: set up and publish cloud environments, start tasks from the web, CLI or GitHub, review diffs, open PRs, and see which plans include it.
Codex config.toml: location, precedence, profiles and key options: Where Codex config.toml lives, which layer wins, how profile files work since 0.134.0, and the options worth setting: model, approvals, sandbox, web search.
codex exec: run Codex non-interactively in scripts and CI: codex exec guide: run Codex without the TUI, pipe input, get JSON or schema-checked output, resume runs, and keep safe sandbox defaults in CI.
Codex GitHub Action: run Codex in your CI workflows: How the Codex GitHub Action works: set up openai/codex-action, write a PR review workflow, pick permission profiles and safety strategies, and use outputs.
Codex MCP: add servers with codex mcp add and config.toml: How to use MCP in Codex: codex mcp add for local and remote servers, the [mcp_servers] tables in config.toml, OAuth login, tool filters and timeouts.
Codex plugins: install, marketplaces and building your own: Codex plugins explained: what a plugin bundles, how to install one with /plugins, add marketplaces with codex plugin marketplace add, and build your own.
Codex sandbox and approvals: modes, network access and presets: Codex sandbox guide: the read-only, workspace-write and full-access modes, approval policies, network access, protected paths, OS enforcement and safe presets.
Codex skills: where they live, how Codex uses them, and sharing them with Claude Code: Codex skills explained: the folders Codex reads, how it picks a skill or runs one with $name, installing skills, and sharing one skill folder with Claude Code.
Codex VS Code extension: install, features, settings and cloud tasks: Codex VS Code extension guide: install in VS Code, Cursor, Windsurf or JetBrains, use editor context, set permissions and settings, and run cloud tasks.

Gemini

Gemini CLI MCP: add, configure and manage MCP servers: Gemini CLI MCP setup: the mcpServers block in settings.json, gemini mcp add and list, remote servers with OAuth, tool filtering, security and working examples.
Gemini CLI skills: where SKILL.md folders go and how to share them: Gemini CLI skills explained: where SKILL.md folders go, how activation works, gemini skills commands, extensions, and sharing skills with Claude Code and Codex.
Gemini CLI: install, sign in, configure and use Google's terminal agent: Gemini CLI guide: install, sign in, what changed for free users in 2026, GEMINI.md, settings.json, MCP servers, extensions, slash commands and headless mode.
GEMINI.md: how Gemini CLI's context file works, with a template: GEMINI.md explained: where Gemini CLI loads it from, the hierarchy, @ imports, /memory commands, using AGENTS.md instead, and how it compares to CLAUDE.md.

Cursor

Cursor Agent: modes, tools and checkpoints explained: How Cursor Agent works: Agent, Ask, Plan and Debug modes, the tools it can call, checkpoints for undoing changes, run modes for approvals, and prompting tips.
Cursor background agents (now Cloud Agents): setup and use: Cursor background agents are now called Cloud Agents. How they work, where to start them, environment.json setup, secrets, hooks, plans and security.
Cursor Bugbot: AI pull request review, setup, rules and pricing: How Cursor Bugbot reviews pull requests: setup on GitHub and GitLab, review commands, BUGBOT.md rules, bugbot.yaml, Autofix, limits and usage-based pricing.
Cursor CLI: install the agent command and run it headless: A practical guide to the Cursor CLI: install the agent command, use Plan and Ask modes, run it headless with -p and --force, set permissions, and add it to CI.
Cursor hooks: hooks.json, events and examples: How Cursor hooks work: hooks.json locations, agent and Tab events, exit codes, matchers, prompt-based hooks, cloud agent support and Claude Code compatibility.
Cursor MCP: set up servers with mcp.json and Customize: Cursor MCP guide: adding MCP servers, mcp.json locations and fields, one-click Marketplace installs, OAuth sign-in, the agent CLI and debugging logs.
Cursor Plan Mode: plan before the agent writes code: How Cursor Plan Mode works: switch with Shift+Tab, /plan or --plan, answer clarifying questions, edit the plan, save it to your workspace and build it.
Cursor pricing: plans, usage pools and what each tier includes: Cursor pricing as of October 2026: Free, Plus, Premium, Super and Ultra plans, weekly usage pools, third-party model costs, Teams seats and legacy Pro plans.
Cursor Tab: how Cursor's AI autocomplete works and how to control it: How Cursor Tab works: accept and reject shortcuts, word-by-word accept, multi-line and cross-file edits, jump-in-file, snoozing, Tab hooks and fixes when it stops.

GitHub Copilot

Copilot coding agent (now Copilot cloud agent): how it works: GitHub Copilot coding agent is now Copilot cloud agent. How to assign issues, use @copilot, set up copilot-setup-steps.yml, plans, limits and security.
GitHub Copilot CLI: install, commands, permissions and tips: How to use GitHub Copilot CLI: install with npm, Homebrew or WinGet, sign in, plan mode, -p scripting, tool approval flags, MCP, custom instructions and skills.
GitHub Copilot custom instructions: copilot-instructions.md, paths and AGENTS.md: Set up GitHub Copilot custom instructions: .github/copilot-instructions.md, path-specific .instructions.md files, AGENTS.md support and reusable prompt files.

More agents

Antigravity CLI: install agy, sign in, skills, plugins and MCP: Antigravity CLI guide from Google's docs: install the agy command, sign in, plan limits, skills, rules, plugins, MCP config and moving over from Gemini CLI.
Cline: install, providers, Plan and Act, rules, skills and MCP: Cline guide from the official docs: install the extension or CLI, pick a provider, use Plan and Act modes, and set up rules, skills and MCP servers.
Devin Desktop (formerly Windsurf): what changed, features, skills and pricing: Devin Desktop is the new name for Windsurf. What changed in the rename, Devin Local, rules, skills, MCP, third-party agents and pricing as of October 2026.
OpenCode: install, providers, config, agents, MCP and skills: OpenCode guide from the official docs: install OpenCode 2, connect providers, write opencode.json, set up agents, MCP servers, skills and AGENTS.md rules.

Compare

AGENTS.md vs Cursor rules: which to use, and how to run both: AGENTS.md vs Cursor rules compared: plain Markdown every agent reads against .mdc rules with globs and rule types. A comparison table and when to use each.
ChatGPT Projects vs Claude Projects: how they differ and which to use: ChatGPT Projects vs Claude Projects compared from official docs: files and knowledge, instructions, memory, sharing, local folders and when to pick each.
Claude Code vs Antigravity: Anthropic's and Google's coding agents compared: Claude Code vs Antigravity compared from official docs: surfaces, models, plans, permissions, rules and skills, and which coding agent fits your work.
Claude Code vs Cline: models, pricing, rules and which to pick: Claude Code vs Cline compared from official docs: models and providers, where each runs, approvals, rules and skills, pricing, and which one fits how you work.
Claude Code vs Cursor: how they differ and when to use each: Claude Code vs Cursor compared: a terminal agent against an AI code editor. How each works, rules and skills, where each fits, and how to use both together.
Claude Code vs GitHub Copilot: agent mode, cloud agent, CLI and pricing compared: Claude Code vs Copilot compared: GitHub Copilot agent mode, cloud agent and CLI against Claude Code, with instructions, skills, pricing and which to pick.
Claude Code vs Windsurf (Devin Desktop): agent against AI IDE in 2026: Claude Code vs Windsurf compared now that Windsurf is Devin Desktop: terminal agent against AI IDE, agents, models, rules, skills, pricing and which to pick.
Claude vs ChatGPT for coding: apps, agents, models and plans compared: Claude vs ChatGPT for coding, from official docs: Claude Code against Codex. Models, coding agents, plans, safety defaults and which one to pick.
Claude vs Gemini for coding: models, coding agents and plans compared: Claude vs Gemini for coding, from official docs only: current models, API prices, coding agents (Claude Code, Antigravity), plans, and how to choose.
CLAUDE.md vs AGENTS.md: which file each tool reads and how to keep one source: CLAUDE.md vs AGENTS.md explained: which coding agents read each file, how Claude Code loads AGENTS.md, and three ways to keep one source of instructions.
Cline vs Copilot: open source agent vs GitHub's AI assistant: Cline vs Copilot compared from official docs: editors, models and API keys, plans and prices, approvals, rules, skills, MCP, cloud agents, and which to pick.
Cline vs Cursor: open source agent or AI code editor?: Cline vs Cursor compared from official docs: an open source coding agent against an AI code editor. Models, pricing, rules, skills, and which to pick.
Codex vs Claude Code: an honest comparison for 2026: Codex vs Claude Code compared: models, where each runs, AGENTS.md vs CLAUDE.md, skills, MCP, sandboxing and approvals, pricing, and how to use both together.
Codex vs Cursor: OpenAI's coding agent against the AI code editor: Codex vs Cursor compared: OpenAI's coding agent against an AI code editor. Where each runs, models, AGENTS.md, skills, sandboxing, pricing, and using both.
Codex vs Gemini CLI: OpenAI's and Google's terminal agents compared: Codex vs Gemini CLI compared from official docs: who can use each in 2026, install, sign-in, sandboxing, AGENTS.md vs GEMINI.md, skills, MCP and scripting.
Copilot vs Codex: GitHub's assistant against OpenAI's coding agent: Copilot vs Codex compared: GitHub Copilot in your IDE and on GitHub against OpenAI Codex in the CLI, IDE, app and cloud. Models, AGENTS.md, skills and pricing.
Cursor rules: .mdc project rules, examples and the old .cursorrules file: How Cursor rules work today: .mdc project rules, the four rule types, examples, what happened to .cursorrules, and reusing rules as Agent Skills.
Cursor vs Antigravity: two agentic coding platforms compared: Cursor vs Antigravity compared from official docs: editors, CLIs, cloud agents, models, plans and prices, rules, skills and MCP, and which one fits your work.
Cursor vs Copilot: the AI editor against GitHub's assistant: Cursor vs Copilot compared: an AI code editor against GitHub Copilot in your IDE and on GitHub. Agents, completions, rules, skills, models, pricing and picks.
Cursor vs Windsurf in 2026: how they compare now that Windsurf is Devin Desktop: Cursor vs Windsurf compared for 2026: Windsurf is now Devin Desktop. Agents, Tab, rules, skills, models and pricing side by side, and which one to pick.
Gemini CLI vs Claude Code: differences, access and the Antigravity CLI switch: Gemini CLI vs Claude Code compared from official docs: who can still use Gemini CLI after the Antigravity CLI switch, context files, skills and pricing.
OpenCode vs Claude Code: models, permissions, skills and which to pick: OpenCode vs Claude Code compared from official docs: models and pricing, default permissions, AGENTS.md, skills and commands, and which one to pick.
OpenCode vs Codex: open source terminal agents compared: OpenCode vs Codex compared from official docs: models and providers, plans, permissions and sandboxing, AGENTS.md, skills, MCP, scripting, and which to pick.
Skills vs MCP: what each one does, when to use which, and how they work together: Skills vs MCP compared honestly: what Agent Skills and MCP servers each do, context cost, setup and security, when to pick one, and how to use both together.
Skills vs plugins in Claude Code: what each one is and which to use: Skills vs plugins in Claude Code: a skill is one SKILL.md folder, a plugin bundles skills, hooks, agents and MCP servers. Comparison and when to use each.
Skills vs prompts: when a reusable prompt is enough and when to make an Agent Skill: Skills vs prompts explained: when a saved prompt with blanks does the job, when an Agent Skill is worth it, what each costs in context, and how to convert one.
Skills vs slash commands in Claude Code: what changed and which to use: Skills vs slash commands in Claude Code: custom commands merged into skills. What still differs, how invocation and arguments work, and which format to pick.
Skills vs subagents in Claude Code: what each does and when to use which: Skills vs subagents in Claude Code: skills add reusable instructions, subagents run work in a separate context. How they differ, when to use each, and examples.

Tools

AI prompt library: how to organize and reuse your prompts (and the tools for it): How to build an AI prompt library: what makes a good one, how to store prompts with variables and tags, and a fair comparison of prompt library tools.

Keep your prompts and skills in one place, in your own GitHub. Install every skill with one command.

Get started

Kipt is an independent product, not affiliated with or endorsed by Anthropic, OpenAI, Cursor or GitHub. Claude, ChatGPT, Codex, Cursor and GitHub are trademarks of their owners.