The Ultimate Free AI Toolkit

Turn any text into stunning visuals in one click.

Paste your ideas or notes and Napkin auto-generates clean diagrams, infographics, and graphics to match. A secret weapon for content, posts, and presentations.

Type anything and watch AI build a live website.

A playground where AI hallucinates fully interactive websites from a prompt or an imaginary URL. Genuinely mind-bending — and a surprising way to spark ideas.

Make any character or photo move and dance.

The engine behind countless viral clips — animate a still image or drop a character into any motion. A cheat code for scroll-stopping video content.

Hollywood-style camera moves and VFX from a prompt.

Generate cinematic video with pro camera motions and effects that normally take a full film crew. Creators are using it to make Reels look like movies.

An AI 'super agent' that does the whole task for you.

Give it a goal and it researches, builds pages, and completes multi-step jobs on its own. Like having an assistant that actually finishes things.

One click installs any AI app on your computer.

An app store for open source AI — video models, voice models, image tools — each a one-click install. The easiest way to run tools like Hunyuan Video and Dia locally, free.

Ramble out loud, get clean written notes back.

Speak your messy thoughts and AudioPen turns them into structured, polished text. The fastest way to capture ideas and draft content on the go.

Turn any video, PDF, or idea into a mind map.

Drop in a YouTube link, document, or prompt and Mapify instantly builds a visual mind map to help you understand and remember it. Great for learning fast.

An invisible AI that sees your screen and feeds you answers.

It watches your screen and listens to your calls, then quietly gives you the answer in real time — only you can see it. Built for sales calls, live demos, and debugging. Check your workplace's rules before running it on calls.

Every frontier AI model in one tab — and it pays you.

Type one prompt and two top models answer side by side — ChatGPT, Claude, Gemini, and hundreds more. Pick the better answer and earn credits you can cash out (up to about $50 a month). The real win is learning which model wins which job.

The AI-powered code editor developers love.

A polished editor with AI baked into every keystroke — autocomplete, chat, and multi-file edits. The most popular on-ramp for building with AI.

An AI editor that plans and builds alongside you.

A clean, agentic IDE that can reason across your whole project and make coordinated changes. A strong free alternative to Cursor.

AI autocomplete inside your editor.

The original AI pair programmer. Suggests the next line — or whole function — as you type, right inside VS Code and others. Free tier available.

Turn a prompt into a polished UI in seconds.

Describe a page or component and v0 generates clean, production-ready UI code you can copy straight into your project. Perfect for fast, good-looking front ends.

Build and run a full app right in your browser.

Prompt it and it spins up a complete web app — front end, back end, and a live preview — with nothing to install. Great for prototypes and MVPs.

Ship a working app by chatting with AI.

A build-by-conversation platform that turns your ideas into a real, deployable app. Popular with non-coders launching their first product.

Type what you want. Watch it build the site live.

A free AI coder in your browser, powered by DeepSeek. Paste any website link and it redesigns the whole thing — then click any part of the page to edit just that part. One-click deploy with a free Hugging Face account.

One prompt in, a working app out.

It handles front end, back end, APIs, database, and Stripe payments, then deploys to a live link in one click. Try cloning an app you know first — you learn the tool with zero design decisions. New accounts get free credits.

Code, run, and deploy from anywhere — no setup.

A full development environment in your browser with an AI agent that can build features for you. Handy for learning and shipping without installing anything.

A free, AI-first home for your projects.

Google's agentic code editor. If you're not comfortable in a terminal, this gives you a clean visual workspace to build and preview with AI. A great free option.

A free, open-source coding agent for VS Code.

Plugs an autonomous AI developer into VS Code — it can read your files, run commands, and build features. Bring your own model key and it's completely free.

OpenAI's agent that codes and controls your computer.

The Codex desktop app writes code, previews docs and slides in a sidebar, generates images, and can run tasks across your apps in the background. Included with a ChatGPT plan.

Design a full app or website UI from a prompt.

Describe the interface you want and Stitch generates polished, editable UI designs in minutes — then export to code or hand off to Claude Code. Free with a Google account.

Google's browser-based AI app builder.

A full cloud development environment (formerly Project IDX) that turns text prompts into working full-stack apps. Like Cursor in your browser, from Google.

Build AI apps and get a free Gemini API key.

Prototype with Gemini, generate whole web apps in Build mode, and grab a free API key to power your own projects. The fastest way into Google's AI stack.

An async agent that fixes and builds while you step away.

Assign it a task or a bug and Jules works in the background on your repo, then opens a pull request with the changes. Coding that happens without you babysitting it.

One API key, 50+ AI models — including free ones.

A single key that routes to dozens of models, with a free tier that includes DeepSeek R1, Kimi, Qwen, and more (look for the models tagged 'free'). Perfect for building agents and automations without paying for tokens while you learn.

Free API access to open models, at blazing speed.

Run DeepSeek, Llama, and Mistral through a free API — the whole point here is raw speed, with answers that come back almost instantly. Great for real-time apps and agents that can’t wait on slow models.

Describe your app and get the front end AND the back end.

A no-code builder that handles the parts others skip — database, emails, analytics, and storage all wired up automatically. Free credits to start. Great for client demos: build the working version in front of them, then charge for the real thing.

Google's free, agent-style coder inside your editor.

Install the extension in VS Code (it works inside Cursor too), sign in with a Google account, and get roughly 180k code completions a month plus chat and agent help — all free. The cheapest real AI coding setup there is: $0.

Describe your app in plain English and Spark builds it.

GitHub and Microsoft’s builder: describe an app, change anything by chatting, and deploy in one click. Every change is saved to a GitHub repo so you can preview and roll back like commits. Part of the paid Copilot tiers — if you already pay for Copilot, you may have it now.

The official, vetted skills library from Anthropic.

The canonical source of Claude Skills — generate Word docs, PDFs, and slides, and more. Start here before installing community skills; these are trusted by default.

Make Claude Code build genuinely beautiful sites.

A design-intelligence skill for Claude Code. Install it, then ask Claude to build — and it applies real UI/UX taste for premium, polished results instead of generic layouts.

A loaded starter kit of agents, skills, and commands.

An open-source Claude Code plugin bundling specialized agents, slash commands, hooks, and pre-built MCP configs. A fast way to supercharge your setup.

Give your AI coder a persistent memory.

A 'Memory Bank' system with five integrated modes (Architect, Code, Debug, Ask, Orchestrator) that share project context across sessions — so your agent stops forgetting.

A skill that finds and installs other skills.

From Vercel Labs — searches the open agent-skills ecosystem and installs the right one for your task automatically. Just ask Claude for what you need.

Turn any folder into a queryable knowledge graph.

An open-source skill that maps your code, docs, and papers into a graph so Claude Code, Codex, Cursor, or Gemini CLI can navigate by structure — not just search text.

Turn Stitch designs into full websites autonomously.

Google Labs' official skills for the Stitch MCP server — including the autonomous 'stitch-loop' that hands designs to Claude Code and builds multi-page sites for you.

The AI assistant that started it all.

OpenAI's flagship chatbot for writing, coding, analysis, images, and voice. The most widely used AI tool in the world, with a capable free tier.

The AI I trust for writing, coding, and reasoning.

Anthropic's assistant, known for thoughtful long-form writing, strong coding, and careful reasoning. Excellent for real work, not just quick answers.

Google's AI, wired into everything you already use.

A powerful multimodal assistant with huge context and deep integration into Google apps. Great for research, docs, and working across images and video.

AI answers with real sources you can check.

A search engine that reads the web and gives you a cited answer instead of ten blue links. The fastest way to research anything and trust the result.

xAI's assistant with a live view of X.

A capable chatbot with real-time access to what's happening on X. Good for current events, trends, and a more unfiltered style.

Frontier-level reasoning, free to use.

An open, high-performance model that rivals the big names at coding and math — available free in the browser. Remarkable value.

Does everything ChatGPT does — for free.

Image generation, coding, even custom agents. The closest thing to a full ChatGPT replacement that costs nothing.

Alibaba's model — strong at reasoning, coding, and math.

A serious free chatbot most people have never opened. Pick one chatbot and use it every day for a week — that teaches you more than hopping between five tools.

Turn your own documents into an expert you can ask.

Upload PDFs, notes, and links and NotebookLM becomes a grounded assistant on your material — it can even generate an audio 'podcast' summary. Free from Google.

Run powerful AI models on your own computer.

A free tool to download and run open models locally — private, offline, and no subscription. Ideal if you care about data privacy or want to experiment.

A friendly desktop app for local AI.

Point-and-click way to run open models on your machine with a clean chat interface. The easiest on-ramp to private, offline AI.

A frontier-level model at a fraction of the cost.

MiniMax's M-series models rival the big names on coding and reasoning — and can even power Claude Code via an Anthropic-compatible endpoint for far less. Great value for heavy use.

An AI browser that thinks in tabs, built on Gemini 3.

Google Labs' experimental browser with 'GenTabs' that research, compare, and build things for you as you browse. Invite-only for now — join the waitlist.

Chat with your open tabs.

Summarize pages, draft emails, ask questions about whatever is on screen. Try it for a normal workday — you'll feel the moment browsing turns into delegating.

50+ AI models in one single HTML file.

An open-source, single-file chat interface that routes through your own API key to reach Claude, GPT, Gemini, Grok, DeepSeek and more — all client-side in your browser. Powerful and unfiltered; use responsibly.

The gold standard for stunning AI art.

Produces the most beautiful, stylized images of any generator. Worth the subscription if visuals are central to your brand or content.

The image AI that actually gets text right.

Best-in-class at rendering readable words in images — logos, posters, ad creatives. Generous free tier.

A creative suite built for consistent visuals.

Great for game art, product shots, and repeatable styles, with fine-grained controls. Daily free credits.

Commercially-safe AI images from Adobe.

Trained on licensed content, so it's safe to use in client work. Integrates directly into Photoshop and the Adobe suite.

Generate logos, icons, and vectors on-brand.

A design-focused AI that outputs editable vectors and keeps a consistent brand style — ideal for real design deliverables, not just pretty pictures.

Design anything, now supercharged with AI.

The all-in-one design tool for social posts, thumbnails, and decks, with AI generation, background removal, and writing built in. Beginner-friendly and free to start.

Where modern interfaces are designed.

The industry-standard tool for designing websites and apps, now with AI features. Design a layout, then hand it to Claude Code to build.

Watch your image form as you type.

Real-time generation and upscaling with a fun, fast canvas. Great for ideation and turning rough sketches into polished visuals.

Millions of assets, plus AI generators in one place.

Stock photos, vectors, and templates alongside AI image and video tools. A one-stop shop for creative assets.

Generate and edit images from a single sentence.

Google's image model built into Gemini — create or edit photos with plain-text instructions and remarkable consistency. Free to use, right inside Gemini.

Turn a single photo into a full 3D model.

Microsoft's open-source model that converts one image into a high-res, textured 3D asset you can export and use in games or design. MIT-licensed and free.

Turn any image into a 3D model.

Tencent's free open source tool. Upload an image, hit Gen Shape for a preview mesh in seconds, then Gen Textured Shape and download the glb — it works everywhere. People sell models like these on Sketchfab and TurboSquid.

Free daily credits. Near-perfect text in images.

ByteDance's image model — it beats most paid models on benchmarks and renders text inside images almost perfectly. Sign in, describe your image, generate. Try a poster or an ad for a brand you like.

A full AI video studio in your browser.

Generate and edit video from text or images, with pro-grade tools for creators and filmmakers. One of the most capable video models available.

Cinematic video from a single prompt.

OpenAI's text-to-video model produces strikingly realistic clips. Included with ChatGPT paid plans — a glimpse of where content is heading.

High-quality AI video with a free tier.

Turns text and images into smooth, detailed video clips. A favorite for social content thanks to its free daily credits.

Cinematic AI video with no duration or usage caps.

Text to video and image to video with camera control, music, sound effects, and lip synced voices in one tool. Fully open source (SkyReels-V2 on GitHub) — a full studio for $0 if you make faceless content.

One image plus a prompt becomes a full video clip.

Tencent's free open source video model. Try it online, or run it on your own NVIDIA GPU with a one-click Pinokio install (it runs in ComfyUI too).

A talking AI avatar of you — no camera needed.

Create realistic spokesperson videos and clone your own avatar and voice. Great for faceless channels and scaling content in any language.

The most realistic AI voices you can generate.

Studio-quality voiceovers and voice cloning in dozens of languages. The go-to for narration, faceless content, and dubbing.

400+ voices, 60 languages, no signup.

Unlimited text-to-speech with no account needed. A free ElevenLabs alternative for reels, YouTube, and audiobooks — check the license page before paid client work.

Nearly 300 realistic voices in 70+ languages.

Unlimited use, completely free. Another strong ElevenLabs alternative for narration and faceless content.

200+ natural voices — it can clone yours too.

Text to speech plus voice cloning in one free tool. Narrate your content without ever recording.

AI voices with real emotion — it acts the script.

It performs full conversations with real tone, not a flat read. Write [S1] and [S2] for two speakers and add (laughs) where you want it — it performs them. Free online, or one-click local install via Pinokio.

Write a full song from a text prompt.

Describe a vibe and Suno composes original music with vocals. Perfect for custom background tracks and jingles you fully own.

Edit video by editing the transcript.

Delete a word from the text and it's gone from the video. Includes filler-word removal, AI voices, and screen recording — editing made effortless.

Turn one long video into 10 viral clips.

Feed it a podcast or YouTube video and it finds the best moments, cuts them into vertical shorts, and adds captions automatically. A huge time-saver.

Add scroll-stopping captions in one click.

Auto-generates animated captions, emojis, and B-roll for short-form video. The polish behind a lot of viral Reels and TikToks.

The free editor built for social video.

A powerful, beginner-friendly editor with AI captions, effects, and templates. The default choice for editing Reels and Shorts on desktop or phone.

A free AI video studio that runs on your own machine.

Lightricks' open-source desktop app pairs a full timeline editor with on-device AI generation — text-to-video and image-to-video with no subscription. Runs locally on supported NVIDIA GPUs.

One prompt into a full YouTube documentary.

An end-to-end platform that researches, scripts, voices, and edits a finished long-form video — matched B-roll, visuals, and music included, on a type-to-edit timeline.

A free, self-hosted alternative to Higgsfield.

Bundles 200+ image and video models (Flux, Kling, Sora, Veo, Seedance, Wan) into one open-source dashboard with dedicated studios. MIT-licensed — run it yourself for free.

Turn long podcasts into captioned short clips, automatically.

Feed it a video and it finds the best moments, cuts vertical clips, and adds subtitles. It also has an API, so you can wire it into an n8n workflow that clips new episodes on a schedule — that exact automation is a real service podcasters pay monthly for.

Hundreds of AI videos, each with the exact JSON prompt.

Find a clip close to what you want, copy its JSON prompt, and swap in your product with ChatGPT — the JSON locks the camera, lighting, and scene so you only change the subject. Same cinematic look, your product, and far fewer wasted credits.

Open-source voice cloning from about 5 seconds of audio.

Resemble AI's free model clones a voice from a short sample and beat a big paid tool in a blind test. Upload a clean sample, type your script, set the emotion, and generate — unlimited, no bill. Source is on GitHub. Only clone voices you have permission to use.

Your notes, docs, and projects — with AI inside.

The all-in-one workspace with AI that can summarize, write, and answer questions across everything you've saved. Where a lot of solo founders run their business.

Make everything you write clear and correct.

Real-time grammar, tone, and clarity fixes everywhere you type, plus AI drafting. Free tier covers the essentials for polished writing.

AI built specifically for marketers.

Generates on-brand ads, emails, and campaigns at scale with brand-voice controls. A strong pick if content is your core business.

Instant copy for ads, emails, and posts.

Templates for every kind of marketing copy, plus automated GTM workflows. Free tier to get moving fast.

Generate beautiful decks and one-pagers from a prompt.

Describe your idea and Gamma builds a polished presentation, document, or web page. Perfect for client proposals and pitches without touching a slide.

Long-form articles and SEO content on autopilot.

Generates blog posts, landing pages, and SEO-optimized content with research built in. Handy for filling out a site or content calendar fast.

Open-source automation you can self-host free.

Connect hundreds of apps and AI models into visual workflows — no per-task fees if you host it yourself. Its new AI chat builder can generate an entire workflow from a sentence or even a screenshot, then edit and debug the ones you already have. The most flexible automation tool for builders.

Visual automation for connecting your whole stack.

A drag-and-drop canvas to link apps and automate multi-step processes. More powerful (and cheaper at scale) than most no-code tools.

Connect 6,000+ apps with zero code.

The easiest way to automate 'when this happens, do that' across your tools. Great starting point for simple, reliable workflows.

Drag-and-drop automations powered by AI.

Build AI-driven workflows on a visual canvas — scraping, summarizing, enriching, and more. Made for AI-first automation.

Build a team of AI agents to run tasks.

Create and orchestrate custom AI 'workers' that handle sales, research, and support. A no-code way to put agents to work in your business.

An AI agent that completes whole tasks for you.

Give it a goal and it plans, browses, and executes end-to-end — building things while you step away. A look at where autonomous AI is going.

Manus power on your own computer, no usage cap.

The free open source alternative to Manus. Clone the repo, pip install the requirements, paste an API key into the config (Claude works great), and run python main.py. Give it one clear task and watch how it plans.

Personal AI assistants that handle email and meetings.

Build AI helpers that manage your inbox, schedule, and follow-ups automatically. Like hiring an assistant that never sleeps.

Post to every social platform on autopilot.

Connect Instagram, TikTok, X, YouTube, LinkedIn and more, then schedule and auto-publish everywhere from one place — with full API access on the free tier. Pairs perfectly with AI content.

Scrape any website and feed it to your AI.

A marketplace of ready-made scrapers (Instagram, Google Maps, Amazon, and thousands more) plus a platform to run your own. The data engine behind serious AI agents.

280+ ready-to-import n8n automations.

The largest free collection of prebuilt n8n workflows — WhatsApp bots, Google Sheets AI analysts, content pipelines and more. Import one and run it in minutes.

Thousands of ready-made n8n automations.

Real systems, not toy demos — download a .json, choose Import from File in n8n, connect your accounts, switch it on. These templates are the exact kind of systems businesses pay for. Learn one deep and you have something to sell.

An infinite agent that keeps working after you close the laptop.

Hand it long, multi-thousand-step jobs — videos, 3D models, data extraction, whole apps — and it runs 24/7 on its own. Give it one big, clearly described task before bed and check the result in the morning. Compare it against OpenManus and keep whichever fits.

n8n plus Lovable in one — AI apps from plain English.

Describe an idea like 'build me an AI ad generator' and Opal builds the whole workflow AND a clean front end to run it — it can even generate video with Veo 3. Edit it visually and share the app in one click. Try rebuilding one thing you do by hand every week.

A free API for stocks, ETFs, and indicators.

Free real-time and historical market data — stocks, ETFs, crypto, and technical indicators. Grab an API key and let Claude build your own custom market dashboard.

Connect Claude Code to your TradingView charts.

An open-source MCP server that lets Claude read your TradingView charts, analyze setups, write Pine Script, and scan watchlists via natural language. For research, not financial advice.

Let Claude pull market data and manage a Bybit account.

Bybit's official MCP server connects Claude (and Cursor/VS Code) to the exchange to fetch live data and execute actions via natural language. Real money — configure carefully and start on testnet.

An open-source AI agent that trades autonomously.

A community-built agent that runs live trades on Hyperliquid with a PnL dashboard. Experimental and high-risk — study the code and paper-trade before ever going live.

An AI agent that funds and replicates itself on-chain.

A wild open-source experiment: an agent that boots its own crypto wallet, earns, and evolves without a human. Fascinating and bleeding-edge — explore it as research, not a get-rich scheme.

Put your site online for free in ~30 seconds.

Connect a GitHub repo and get a live URL with easy custom domains. The fastest way to deploy modern websites — this platform runs on it.

Simple, free hosting for any website.

Drag-and-drop or Git-connected deploys with forms, functions, and a generous free plan. A great alternative to Vercel.

A full backend — database, auth, storage — free.

An open-source Firebase alternative that gives your app a Postgres database, user login, and file storage in minutes. Powers this platform too.

Speed, security, and cheap domains.

Free CDN and protection to make any site fast and safe, plus at-cost domain registration. Essential infrastructure you'll use again and again.

Accept payments online, the developer's way.

The standard for taking cards, subscriptions, and checkouts. No monthly fee — you only pay a small cut per transaction. This is how you get paid.

Send beautiful transactional emails from your app.

A clean, developer-friendly email API for receipts, welcome emails, and notifications. Generous free tier to start.

Where your code lives in the cloud.

Free storage and version history for every project, and the bridge that auto-publishes your site to Vercel on each change. Non-negotiable for building.

Build your own everything, from scratch.

A massive collection of step-by-step tutorials for recreating the tech you use — your own database, git, browser, and more. The best way to truly learn.

Hundreds of free APIs for your projects.

A giant, categorized list of free public APIs — weather, finance, games, AI, and more. Fuel for building useful apps fast.

A goldmine of proven AI prompts.

Thousands of ready-to-use prompts that turn ChatGPT into any expert you need. Copy, paste, and get better results instantly.

6,000+ lines of real prompts from billion dollar AI tools.

The actual system prompts behind Cursor, Manus, Bolt, Lovable, Devin, and more. Read one tool's prompt top to bottom, notice the patterns — role, rules, tools, guardrails — then steal the structure for your own prompts.

Official recipes for building with Claude.

Real code examples for getting the most out of Claude — tool use, RAG, agents, and more. Perfect for going from user to builder.

Learn how large systems are actually built.

A famous, free guide to designing scalable systems — with diagrams and interview prep. Essential if you want to build serious products.

Visual roadmaps for any tech career.

Clear, community-built paths showing exactly what to learn and in what order — front end, back end, AI, and more. Your free curriculum.

Learn to code for free, from zero.

Thousands of hours of hands-on lessons and certifications in web development, Python, data, and AI — all 100% free. A proven place to start.

A complete, free path to becoming a developer.

A full open-source curriculum that takes you from beginner to job-ready with real projects. Free forever.

Learn probability and statistics, visually.

A gorgeous free site from Brown University that teaches stats through interactive, hands-on simulations. The reference behind those 'visual ML' website builds.

Hundreds of interactive science experiments.

University of Colorado's free, browser-based physics, chemistry, and math simulations. Perfect inspiration (and reference) for building interactive learning sites with AI.

A free AI teacher that watches your screen and helps.

An open-source assistant that lives next to your cursor, sees what you're doing, talks to you, and points at the exact button you need. Learn any app hands-on.

1,001 real AI use cases from the world's top companies.

Google Cloud's free, constantly-updated list of how Fortune 500s actually use AI — organized by function. A goldmine for business ideas and offers you can sell.

Stop letting ChatGPT be your yes-man.

A pack of custom instructions that turns ChatGPT into a brutally honest mentor — it pressure-tests every idea, names the biggest flaw, and forces your thinking to be bulletproof. Paste it once into your settings.

Apply to hundreds of jobs on autopilot.

A 4-prompt workflow that analyzes your resume, rebuilds it ATS-proof, scrapes matching jobs, and auto-applies with tailored resumes and cover notes using ChatGPT Agent Mode.

Turn one selfie into a $300 headshot.

Upload a selfie, paste this prompt with thinking mode on, and get an ultra-realistic LinkedIn-ready studio headshot — with the exact lighting, wardrobe, and framing spelled out.

Make ChatGPT scrub you off the internet.

A 4-prompt sequence that finds every data broker exposing your info, writes the removal requests, submits them via Agent Mode, and re-checks weekly on a schedule.

Let Claude interview you and build its own context file.

Instead of writing a CLAUDE.md by hand, this prompt makes Claude ask you 40–50 smart questions, then compiles a clean context file so every future session already knows how you work.

One structure for almost any task.

Goal: [what you're working on and who it's for] Request: [your ask in one sentence] Output: [exactly how you want it] Constraints: [what must not happen] Checkpoints: pause only for destructive actions or real scope changes. Otherwise, keep going and report when done.

Fix a module and don't stop until tests pass.

/loop 30m — refactor the auth module until every test passes. Commit when it's green.

A health report on your whole codebase.

/loop 1h — audit this repo and produce a full health report: risks, dead code, quick wins, and a prioritized fix list.

Ship a working first version, milestone by milestone.

/loop 2h — build a working MVP for this spec. Commit each milestone and report what's done at every step.

A big migration, run start to finish.

/loop 4h — migrate the codebase to TypeScript strict mode. Fix types file by file, keep the build passing, commit as you go.

Stop it interrupting you for things you already approved.

Paste this into every long run: Pause only for: 1. Destructive or irreversible actions 2. Real scope changes outside the brief 3. Input only I can provide Otherwise, keep going and report when done.

Make it remember lessons across runs.

Keep a lessons.md file. One lesson per entry: # Auth fix — avoid the JWT refresh loop lesson: always invalidate the old token before issuing a new one # Migration — batch the writes confirmed: breaks past 500 rows without batching

Match the effort level to the job.

Low / Med — quick questions, basic research. Fast and cheap. High — the default. Best quality for the cost on most tasks. Xhigh — complex builds and multi-step analysis. Full capability. Ultracode — autonomous orchestration across multiple agents.

Turn ideas into memorable hand-drawn diagrams.

Generates hand-drawn diagrams and flowcharts with animated visual aesthetics.

A full studio for cinematic motion and 3D.

Massive toolkit for motion graphics, 3D scenes, animations, and interactive visuals.

Sleek monochrome interfaces on demand.

Creates minimal interfaces inspired by Nothing Phone's industrial design language.

Advanced image generation with reasoning.

Image generation with reasoning modes, editing tools, and platform-specific presets.

Polished posters and graphics, fast.

Creates posters, graphics, and artistic visuals using modern aesthetic principles.

Procedural art from code.

Produces unique procedural artwork using particles, flow fields, and seeded randomness.

A pro design review for any interface.

Audits interfaces against real design rules for typography, spacing, and accessibility.

Direct-response copy that converts.

Advanced marketing toolkit for headlines, offers, objections, and conversion messaging.

Thumbnail prompts built to get clicks.

Generates optimized thumbnail prompts designed to maximize clicks and curiosity.

Scroll-stopping hooks on tap.

Creates powerful hooks using proven marketing and storytelling frameworks.

Kill the robotic AI voice.

Removes robotic AI writing patterns so content reads naturally human-written.

Turn anything into podcasts and quizzes.

Converts articles, PDFs, videos, and sites into podcasts, quizzes, and presentations.

Sharp, timeless writing with zero fluff.

Produces clean prose without filler or obvious AI writing patterns.

Run your whole X workflow.

Posting, monitoring, extracting, scheduling, and managing content on X/Twitter.

Long-form straight to X.

Publishes long-form articles directly to X with proper formatting and threading.

Color science for perfect palettes.

A color system for palettes, accessibility, contrast, and visual harmony.

Bold UI, never generic.

Forces bold modern UI decisions before coding and kills generic AI interfaces.

Your entire content engine.

A full viral-content workflow for posts, reels, thumbnails, analytics, and growth.

Find hidden connections in your notes.

Surfaces hidden connections in your knowledge base via autonomous brainstorming.

Code cinematic videos with React.

Creates videos programmatically with React animations, captions, effects, and rendering.

A self-improving research loop.

A workflow that continuously modifies, verifies, and optimizes research tasks.

Content that sounds like you.

Learns your writing voice and creates consistent, non-generic content.

Steal viral reels, in your voice.

Reverse-engineers viral reels and rewrites them in your own content style.

Score a post before you publish.

Scores social drafts against your past performance data before publishing.

Read and build PDFs.

Anthropic's official skill to extract text, tables, and metadata and manipulate PDFs.

Word docs, done right.

Anthropic's official skill to create, edit, and analyze Word documents.

Spreadsheets with real formulas.

Anthropic's official skill for Excel operations including formulas and charts.

Build decks from a prompt.

Anthropic's official skill to create PowerPoint presentations with templates.

Custom GIFs for Slack.

Anthropic's official skill to generate custom GIFs for team communication.

Keep every message on-brand.

Anthropic's official skill to maintain brand voice and messaging consistency.

Memos and announcements, fast.

Anthropic's official skill to draft internal memos and team announcements.

Give Claude new powers.

Builds custom MCP servers to extend Claude with your own tools and data.

Ship React artifacts instantly.

Builds interactive React artifacts styled with Tailwind CSS.

A skill that builds skills.

Teaches the method for developing effective, high-quality Claude skills.

Real browser tests, automated.

Playwright-based UI testing that drives your web app like a user.

Red, green, refactor.

Writes code with a disciplined RED-GREEN-REFACTOR workflow.

Find the real bug, not a symptom.

A 4-phase root-cause analysis that hunts bugs methodically.

Trace a bug to its source.

Deep problem investigation with full dependency analysis.

Prove the fix before you ship.

Validates fixes and behavior before marking work complete.

Layered error handling.

Adds multiple validation layers so failures get caught early.

Stop writing bad tests.

Identifies and fixes the most common testing mistakes.

Two features, zero context-switching.

Work on multiple branches at once without stashing or switching.

Reviews that go faster.

Preps clean, formatted diffs so your PR gets reviewed quickly.

Turn feedback into better code.

Integrates constructive PR feedback the right way.

Clean merges, clean history.

Guides merge decisions and keeps git history tidy.

Sharpen an idea with the right questions.

Socratic design refinement that pressure-tests your thinking.

A real plan before you build.

Creates detailed, step-by-step implementation strategies.

Run the plan, checkpoint by checkpoint.

Batch execution with progress checkpoints along the way.

Quality-gated builds for big tasks.

Iterates on complex work with quality gates between steps.

Kill flaky async tests.

Async testing patterns that wait on conditions instead of timers.