Trending AI projects this week, explained

Follow and direct coding agents on your computer through web, desktop, and mobile interfaces.

For developers building computer-use agents: agent-desktop lets an AI agent read and operate desktop apps on Mac, Windows or Linux from a command line.

A skill you install into an AI coding agent so that "audit this codebase" runs a structured hunt for vulnerabilities rather than one opinionated read.

Matt Pocock’s own agent skills, straight from his .agents directory: small, composable skills for doing real engineering work with AI coding assistants rather than vibe coding.

Help AI agents answer questions about company databases using reviewed definitions and SQL, with each answer showing its steps.

Pull text, links and data from web pages or whole sites into files.

Impeccable is a design skill for AI coding agents that pushes back against the handful of visual habits (the same font, the same gradient, cards nested in cards) that show up on nearly every AI-generated interface.

A plugin that records what an AI coding agent did, turns those observations into searchable summaries, and brings the relevant decisions and fixes into a later session.

Chat with a roster of AI bots that each have their own model, computer and apps, and approve risky actions.

Have your coding assistant build or audit web pages that avoid the usual AI look.

Turn Claude Code into an organized game team that plans, designs, codes and reviews your game with you.

Install a ready-made routine so your AI coding agent plans before it builds, tests its changes and reviews its own work from a fresh context.

Describe what you want to build and have your AI coding agent ask questions, agree a written design with you, and only then write code from a step-by-step plan with tests.

Spec Kit gives your AI coding agent a step-by-step process to build a feature, fix a bug or assess an idea.

An add-on for AI coding assistants that maps code, docs, PDFs and images into a graph you can ask questions about instead of searching files.

Read documents with page-by-page citations, practice daily and get video lessons from an AI tutor.

Send AI requests to one local server that routes across many providers' free tiers, for developers and coding-agent users who want to stack free quotas.

An inference engine written in plain C that tries to run very large models — hundreds of billions to trillions of parameters — on hardware that could not normally hold them.

A curated index of step-by-step guides for recreating the technologies you use every day.

Paste in text that sounds machine-written and get back a version that reads as if a person wrote it, with the same meaning.

Feed in past open, high, low and close prices and get a forecast of the next candles.

Add ready-made instruction folders so Claude handles tasks like PDFs, slides and brand-styled documents the same way each time.

Open web pages without a screen, turn them into markdown or data, and drive them from scripts or an AI agent.

Learn to build AI apps through example projects, from reading text in images to chatting with your documents.

Review code, test in a real browser and ship code changes with typed commands in your coding agent.

A command-line tool for Git worktrees, aimed at people running several AI coding agents at once.

Ask your coding agent to make a promo video for your product and get back a finished MP4 you can keep editing.

Handy lets you press a shortcut, speak, and see your words appear in any text field, transcribed on your own computer.

Give Orkas a goal and its AI agents plan, research, write and build the work on your computer.

Have an AI agent review screens, fix accessibility, refine type and colors, and build pages from designs.

Get text answers from the huge Kimi K3 AI model using one CPU and no graphics card.

A skill, plus an optional proxy, that makes an AI coding agent write in clipped fragments instead of full explanatory prose.

Give an AI agent a real browser it can control with short commands: open pages, read them, click and fill forms.

Give your AI coding agent ready-made marketing know-how for copywriting, SEO, ads and A/B tests.

Compare hosted services with free tiers, from web hosting to email.

Give an AI coworker a job, connect the tools it needs and watch it work.

Install ready-made skills so Claude Code, Cursor or Codex can make short videos, web pages and articles.

Find and install reusable instructions for coding agents such as Claude Code, Codex, or Cursor, or create a skill for a repeated task.

Project NOMAD turns a small computer into a self-contained library and AI assistant that works with no internet connection: offline Wikipedia and reference books, Khan Academy-style courses with progress tracking, downloadable regional maps, and a local AI chatbot you can hand your own documents to.

Continue your DeepSeek Harness sessions from a phone, another computer or a browser while the work keeps running on your own machine.

A free hands-on tutorial series on large language models, grown out of two courses at Shanghai Jiao Tong University on frontier language technology and AI security.

Multica is a project-management workspace where AI coding agents get assigned issues the same way a human teammate would: they pick up the work, comment on progress, raise blockers, and hand it back for review, all visible on the same board as your human team.

Learn how AI agents plan, call tools, remember context and work together, one lesson at a time.

Have tests, scripts or an AI agent open web pages, click, type and check results in Chromium, Firefox and WebKit.

Upload documents and ask questions with traceable citations.

Learn to build apps with generative AI, from writing good prompts to making chat, search and image-generation apps.

Enter a domain to see how AI models describe your product and which competitors they name.

Build React app interfaces with 150+ ready-made components and themes.

Cut videos and add auto captions, voices and effects on your own computer.

Compress older coding-agent messages into searchable summaries and restore exact context later.

Learn to build AI systems yourself, from the math behind models to retrieval pipelines and agents, by writing and running real code.

A local voice studio that covers both halves of speech: going out and coming in.

Set up a service that creates and manages separate Kubernetes clusters on your own NVIDIA GPU servers.

Pose characters, set up a 3D scene and camera, then generate images or video from it.

Browse and copy community prompts that work with ChatGPT, Claude, Gemini and others.

json-render is a programming toolkit (framework) for "generative UI": instead of letting a model write arbitrary code for a screen, you give it a fixed catalog of components it is allowed to use, and it returns structured JSON describing which ones to show and how, which the framework then renders.

Send sign-ups, logins, orders and payouts and get an approve, step up, review or block decision with reasons.

Stroq scans your recent Claude Code and Codex sessions for keys that left your machine.

Describe a website task and Pomerado builds a reusable tool your AI assistant can call, including tasks behind login or two-factor authentication.

PGX creates isolated, short-lived PostgreSQL-compatible databases for coding agents, CI jobs, pull requests and tests.

Open-source agent skills and a small CLI that help with newsjacking, pitches, journalist fit checks, and AI-visibility writing once installed into a supported agent.

Lets developers have their coding agent design a landing page or dashboard with an art direction, real pictures and a browser check.

Idol turns one photo, render or generated image of an object into a 3D Gaussian splat file.

Describe a scene or attach a picture and get a single web page that draws it in interactive 3D.

LTX-2.5 is a 22-billion-parameter open-weights model from Lightricks that generates video and sound together: describe a scene in words, or hand it a photo, and it produces a clip with a soundtrack in one pass, so picture and audio stay in sync.

Run a 27-billion-parameter model that reads text, images and video and writes text back, on hardware you control.

Transcribe recordings up to an hour long in 40 languages, with speaker labels and timestamps, or translate speech.

Create images from text or edit pictures using reference images inside ComfyUI, for people who make images and run ComfyUI themselves.

Turn short text requests into 8-megapixel images and edit pictures on your own graphics card.

Use its Python encoder to format text and image chats, then run the model and get written replies.

Chat with a small text model that can call tools and give structured answers, for developers building local agents or specialized apps.

Give a model up to a million tokens of text and images in one go, such as a long agent session or a large pile of input, and get text back.

Viggle’s turbo build is a distilled, faster version of Qwen-Image-2.1, the open image model: where the original takes 40 passes to render a picture, this one takes about six, which Viggle reports makes it roughly five times faster from prompt to finished image.

Unsloth’s release repackages Qwen-Image-2.1 — the open image-generation model — into GGUF files, the compact format that runs on ordinary computers through tools like llama.cpp and Stable Diffusion.cpp instead of demanding a datacenter GPU.