Helps your AI agent design web pages and app screens that look made for your project, not like a template. It guides choices about colors, fonts, layout, motion and wording.
Helps you create new skills for Claude, improve existing ones, and test whether they work. It also tunes a skill's description so Claude uses the skill at the right times.
Makes your AI agent find out what you really want, and why, before it builds anything. It works for software features and for other plans, like a talk, a business or a renovation.
A skill that makes your AI agent find the real cause of a bug before it changes any code. It follows four fixed steps, so the agent stops guessing and piling up quick fixes.
A skill that makes Claude write a test before any new code, check that the test fails, and then write just enough code to pass it. This catches bugs early and shows that each test checks the right thing.
Carries out a written development plan for you by giving each task to a separate helper agent and checking its work before moving on. It ends with one review of all the changes together.
Lets Claude create, edit, and read PowerPoint files (.pptx and .potx). It can build new slide decks, update existing decks and templates, and pull text out of presentations.
A skill that makes your AI agent look for a matching skill before it answers or acts, even before it asks you a question. It is the starting rule for the Superpowers skill set.
Helps Claude split two or more unrelated problems, like separate failing tests, across helper agents that work at the same time. Claude then checks and combines their fixes.
This skill helps you write and fix agent skills by testing them first. You check how an agent fails without the skill, then write the skill until the agent follows it.
This skill helps an agent safely finish a coding branch after the work is done. It checks that tests pass, asks whether to merge, open a pull request, or keep the branch, and cleans up.
Has Claude hand finished work to a separate reviewer agent before you move on or merge. The reviewer sorts problems by how serious they are, so you know what to fix first.
Showing 30 of 150519 results
Helps your AI agent find and install ready-made skills when you ask how to do something. It checks how popular and trusted a skill is before it suggests one.
Lets your AI agent design mobile app screens with Sleek (sleek.design) from a short description. It can also edit your Sleek projects and turn the designs into React Native or HTML code.
A command-line tool that lets your AI agent control a Chrome browser. It can open pages, fill forms, click buttons, take screenshots and collect data from websites for you.
A skill that has the agent question you hard about a plan or design until the weak spots are clear. While it does that, it writes down your decisions and key terms as documents.
Finds parts of your code that are hard to change or test and shows ideas to fix them in a visual HTML report. You pick one idea, and the agent works through the design with you.
A one-time setup that prepares your repo for Matt Pocock's engineering skills. It records where your issues live, which labels to use, and where project docs go.
Helps your AI agent design web pages and app screens that look made for your project, not like a template. It guides choices about colors, fonts, layout, motion and wording.
Generates AI videos from text, images or reference photos with 40+ models such as Veo 3.1, Seedance 2.0, HappyHorse, Wan 2.5 and Grok via the inference.sh belt CLI. It also handles lipsync avatars, video editing, upscaling, foley sound and clip merging.
Generates and edits AI images with 50+ models including GPT-Image-2, FLUX, Gemini 3 Pro, Grok Imagine, Seedream and Reve via the inference.sh belt CLI. It supports text-to-image, editing, inpainting, LoRA styles, text rendering, upscaling and image stitching.
Comprehensive animation skill for HyperFrames providing atomic motion rules, multi-phase scene blueprints, scene transitions, motion-design techniques, and seven runtime adapters including GSAP, Lottie, Three.js, Anime.js, CSS keyframes, Web Animations API, and TypeGPU.
A skill that gives Claude shared words and rules for designing pieces of code that are simple to use and easy to test. Use it when you plan or clean up how your code is split.
Generates and edits AI images with 50+ models including GPT-Image-2, FLUX, Gemini 3 Pro, Grok Imagine, Seedream and Reve via the inference.sh belt CLI. It supports text-to-image, editing, inpainting, LoRA styles, text rendering, upscaling and image stitching.
Generates AI videos from text, images or reference photos with 40+ models such as Veo 3.1, Seedance 2.0, HappyHorse, Wan 2.5 and Grok via the inference.sh belt CLI. It also handles lipsync avatars, video editing, upscaling, foley sound and clip merging.
Gives your AI agent a step-by-step method for hard bugs and slow code. It first builds a quick check that shows the bug, then finds the cause, fixes it and cleans up.
The foundational technical contract for HyperFrames HTML video compositions, defining data attributes, clip/track structure, sub-compositions, variables, media playback, deterministic rendering rules, and project validation.
Gives your AI agent about 450 ready-made video effects for HyperFrames, such as transitions, titles, captions, charts and camera moves. It installs them as code you can edit, so you do not build each effect by hand.
A set of 70 performance rules from Vercel for React and Next.js apps, ranked by how much they matter. Claude follows them when it writes, reviews or cleans up your code, to help your pages load and respond faster.
Non-animation creative direction for HyperFrames videos covering design specs, color palettes, typography, narration scripting, beat planning, audio-reactive visuals, composition patterns, and brand/style decisions.
Generates and edits AI images with 50+ models including GPT-Image-2, FLUX, Gemini 3 Pro, Grok Imagine, Seedream and Reve via the inference.sh belt CLI. It supports text-to-image, editing, inpainting, LoRA styles, text rendering, upscaling and image stitching.
Generates AI videos through the inference.sh CLI (belt) with 40+ models including Veo 3.1, Seedance 2.0, HappyHorse, Wan 2.5 and Grok Imagine Video. Covers text-to-video, image-to-video, reference-to-video, editing, lipsync avatars, upscaling and foley sound.
Checks your website's interface code against Vercel's Web Interface Guidelines and lists each problem with its file and line. It looks at areas such as accessibility, design and user experience.
Guides your AI assistant when you make videos with Remotion, a tool that builds videos from React code. It finds the right guide for each task and shows you a live preview before exporting anything.