Tools Bench.

Product launches and open-source repos with enough signal to earn a second look.

Last Brew Time: Jul 5, 2026, 5:29 AM PT

Insight

The real build energy this run is the pit crew around coding agents, not the agents themselves

Featured

GitHub136.1K

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Market Signal

Why It Has Market Pull

Claude Code is Anthropic's agentic coding tool that runs in the terminal, understands a codebase, and automates routine dev tasks via natural language. It is a category-defining product from a major AI lab with one of the largest followings of any developer tool on GitHub.

  • 136,157 verified GitHub stars and 21,899 forks (repo created February 2025) — among the most-starred dev tools
  • Backed by Anthropic, a top-tier AI lab — a fully-resourced commercial product, not a side project
  • Actively maintained with ~9,700 open issues reflecting massive real-world usage and engagement
  • Enormous educational ecosystem: multiple tutorials each exceeding 1M YouTube views
  • Broad third-party integration momentum (referenced as a supported client across other agent-skill projects)

feedbacks

What People Are Saying

  • "In this course you'll learn how to harness the power of Claude Code in your workflow — install, setup, context, MCP servers, and subagents."YouTube (Net Ninja)

  • "Andrej Karpathy just 10x'd everyone's Claude Code."YouTube video title

  • "800+ hours of learning Claude Code in 8 minutes — newest model, unknown tricks."YouTube video title

  • "Claude Code just changed YouTube forever."YouTube video title

  • "GitHub Copilot CLI, Visual Studio, VS Code, Claude Code and other agents support the skills spec.".NET Blog

  • "Claude Code tutorial for beginners."YouTube (Kevin Stratvert)

GitHub105.8K

High performance self-hosted photo and video management solution.

Market Signal

Why It Has Market Pull

Immich is the category-defining self-hosted, privacy-first alternative to Google Photos and iCloud Photos, with on-device AI features (CLIP natural-language smart search and facial recognition). It is one of the most-starred open-source projects, actively developed daily, with institutional backing and mature cross-platform apps.

  • 105,862 GitHub stars and 6,043 forks, created 2022-02-03, pushed 2026-07-05 (developed today).
  • ~463 active contributors and a near-weekly release cadence; first stable v2.0 in Oct 2025, v3.0 by July 2026.
  • Backed by FUTO since May 2024 (core team full-time) with a preserved AGPL license and a no-paywall commitment.
  • iOS app rated 4.8 stars from ~720 ratings; also on Android and F-Droid.
  • On-device CLIP smart search ('beach sunset 2023') and facial recognition, with repeated Hacker News front-page threads (500–1,400+ points).

feedbacks

What People Are Saying

  • "I've been self hosting immich for the past 6 months. It's pretty awesome. The web is flawless."HN comment

  • "Everything else looked quite solid to me, but face recognition is nowhere near as good as Google Photos."HN comment

  • "I have my photos on an encrypted HDD in a server in my basement, and connect using WireGuard."HN comment

  • "This is quickly becoming a true google photos replacement, and it's all open source on your own server."App Store review

  • "My wife and I back up and share photos automatically without paying a subscription or storing data on other people's servers."App Store review

  • "The easiest to use and most powerful Google Photos alternative."Android Authority review

  • "The app is designed to only exclude photos it backs up rather than also excluding restored ones you download."App Store review

GitHub23.5K

JavaScript in-page GUI agent. Control web interfaces with natural language.

Market Signal

Why It Has Market Pull

page-agent is an official Alibaba open-source JavaScript library that embeds a GUI agent directly inside a web page: it reads the live DOM as text and drives interactive elements from natural language, with no screenshots, headless browser or server required. It launched with a genuinely strong Hacker News reception and has sustained active development.

  • 23,531 stars and 2,032 forks, created 2025-09-23, actively pushed 2026-07-03 with 34 releases (v1.11.0).
  • Show HN drew 147 points and 76 comments (2026-03-05) — a real front-page launch, not a botted spike.
  • Official Alibaba org repo, MIT-licensed, published by an Alibaba engineer.
  • Bring-your-own-LLM via any OpenAI-compatible endpoint (GPT, Claude, Qwen, Ollama); claims ~10x fewer tokens than screenshot approaches.
  • Optional Chrome extension and beta MCP server support extend it to multi-page tasks.

feedbacks

What People Are Saying

  • "Very cool! I'm particularly impressed by the bookmark trick to install it on a page — awesome UX for people to try out the tool."HN comment

  • "i tested the chrome extension, it worked great - i asked it to change the light/dark mode and it navigated to settings, clicked a few tabs, scrolled and found the toggle."HN comment

  • "The PageAgent has access to the security tokens of the currently logged in user. What prevents it from being exploited to send these elsewhere?"HN comment

  • "I don't think building on browser-use is the way to go, it was the worst performing harness of all we tested."HN comment

  • "The biggest challenge with any in-page tool is the tension between deep DOM access and isolation... you will hit CSP issues quickly."HN comment

  • "We need more projects like this in the JS ecosystem."HN comment

GitHub16.0K

Show usage stats for OpenAI Codex and Claude Code, without having to login.

Market Signal

Why It Has Market Pull

CodexBar is a fast-rising macOS menu-bar app that tracks usage, quota and reset windows for AI coding providers — Codex, Claude Code, Cursor, Gemini, Copilot and 50+ others. Built and heavily promoted by a well-known founder-engineer, it solves an acute, current developer pain and ships new releases almost daily.

  • 15,971 stars and 1,339 forks in ~7.5 months, created 2025-11-16, pushed 2026-07-05.
  • ~245 contributors and ~82 releases (1–2 per week) — unusually active for a young menu-bar app.
  • Built by Peter Steinberger (PSPDFKit founder, now at OpenAI) with strong X amplification.
  • Distributed via Homebrew cask, Arch AUR and GitHub Releases with Sparkle auto-updates and a bundled CLI.
  • Privacy-first: reuses existing on-device sessions/keys, no passwords stored; macOS 14+, MIT.

feedbacks

What People Are Saying

  • "CodexBar covers a lot more ground than the minimal Claude-only alternatives."HN comment

  • "This is a great and useful idea for avoiding having to monitor Claude's consumption. I've often exceeded the limit mid-process."HN comment

  • "I was trying to figure out a method to accurately estimate my remaining usage for the five hour sessions."HN comment

  • "CodexBar 0.19.0 is out — Alibaba Coding Plan support, subscription history charts, Cursor dashboard alignment, big Claude stability pass."X post

  • "I really want to get a new release out, just slightly overwhelmed and don't wanna leave everyone hanging."X post

  • "Differing usage between the app menu and real usage in claude monitor."GitHub issue

HF Spaces3.5K likes

Z Image Turbo is a Hugging Face Space tagged with gradio, mcp-server, region:us. It has 3498 likes on Hugging Face.

Market Signal

Why It Has Market Pull

Z-Image-Turbo is a notable, top-ranked open-weights text-to-image model from Alibaba's Tongyi Lab (Tongyi-MAI), a distilled 6B S3-DiT doing sub-second 8-step inference in 16GB VRAM. The scraped row is a well-known HF community member's demo Space wrapping the official model, and the underlying release has strong, verified traction.

  • Official model Tongyi-MAI/Z-Image-Turbo has ~4.93k HF likes and ~921,675 downloads last month (500,000+ on day one).
  • The mrfakename demo Space has ~3.5k likes, matching the scraped 3,498 figure.
  • Ranked #1 open-weights text-to-image model on the Artificial Analysis Image Arena, beating FLUX.2 [dev], HunyuanImage 3.0 and Qwen-Image.
  • GitHub Tongyi-MAI/Z-Image has ~11.7k stars; backed by arXiv paper 2511.22699 and Apache-2.0 license.
  • Efficient at just 6B params (vs FLUX.2 Dev ~32B), running 400-800ms at 1024px on an H800.

feedbacks

What People Are Saying

  • "Z-Image-Turbo - turbo is the right word for mobile"Hugging Face

  • "Is an open-source release for z-image-edit still a possibility? (27 hearts of demand)"Hugging Face

  • "Z-Image Turbo is the new #1 open weights Text to Image model, surpassing FLUX.2 [dev], HunyuanImage 3.0 and Qwen-Image."X

  • "Seeing Z-Image Turbo ranked as the top open-weights model by @ArtificialAnlys is a proud moment for the Alibaba Tongyi-MAI team."X

  • "much faster than Qwen Image or Flux.2 Dev, yet produces comparable, if not better, results."blog

  • "the best open-source image model right now, startling, 400-800 ms at 1024px"review aggregation

Sources

GitHub

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop

Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT 5.5 Thinking, GPT 5.5 Instant, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.

A hand-picked collection of the finest of resources for the most awesome of agents, Claude Code, the undisputed champion of coding companions, from the unstoppable team at Anthropic PBC. A delectable showcase of top tier skills, ambidextrous agents, scintillating status lines, top notch developer tooling, and also we have plugins

Open-source AI penetration testing tool to find and fix your app’s vulnerabilities.

Product Hunt

385

Vida is an AI that learns how you work, remembers what matters, and becomes more like you over time. The more you use Vida, the more it understands your habits, your projects, and your way of getting things done. Eventually, it works like a second version of you—quietly handling repetitive work in the background before you even ask. Today, we’re launching our first 5 SOTA use cases: Reply Rescue · Prompt Rescue · Resume Rescue · Workspace Cleanup · Daily Wrap 95 more to conquer in public.

Type a prompt, get a beautiful planner. Choose a theme that feels like you, then download as a PDF. Built for weddings, hajj, new babies, big moves, and everything in between.

You already own a StreamDeck. It's your iPhone. PhoneDeck turns your iPhone into a fully customizable Mac controller, media controls, app launcher, custom buttons, all on your phone's screen. No hardware. No subscription. Free. Built by a solo developer in 7 days because I needed it myself. Works out of the box. No setup headaches. Download Now: Phonedeck.io

I built CentryAI because I have ADHD and was paying for 11 subscriptions I hadn't used in months. Most trackers make you enter everything manually — that doesn't work if you've forgotten what you're paying for. CentryAI scans your Gmail or iCloud, finds every recurring charge, and scores which ones you're not using. The real pain is cancelling. CentryAI's Cancel Finder locates the exact cancellation page in one tap. No bank linking. Emails are never stored. Available in 18 languages.

The Termi Protocol is a 3D simulation of AI agent workflows. Give your coding agents a face, a desk and a living room. Watch them read, write and run commands live in 3D, like a game. You run the agents; we visualize the process.

Add beautiful backgrounds, realistic frames, stickers, annotations, and more to your screenshots with an easy to use editor. Create eye-catching visuals that grab attention, boost engagement, and make your posts, products, and content stand out instantly.

Hacker News

Zero-Knowledge Proofs (ZKPs) let an untrusted proved show that computation was executed correctly without revealing the inputs to the verifier. However to prove anything, the computation first has to be expressed as a circuit: a system of polynomial equations (constraints) over a finite field. Circuits are the assembly language of zk and every constraint costs prover (and sometimes verifier) time, so production circuits are aggressively hand-optimized. Over the last months, we have been experime... (69 points, 12 comments).

We're a team of three friends who have been working with forms and Open Source for a decade, and we joined forces together to create something where we can apply all of our experience. We recently released GolemUI, an Open Source library to generate forms dynamically from JSON definitions, with a typed layer to simplify authoring. This library has a lot to offer. These are the main characteristics: 1. A JSON engine. The form is governed by a JSON definition that you can store in a DB, version, d... (52 points, 71 comments).

Hey HN, A few months ago, I tried to automate some of my work with the popular AI agent OpenClaw, and then I quickly realized how difficult it is to get it to work with APIs and third-party services securely, which is essential for a lot of work-related tasks. Then I realized OpenClaw is more of a personal assistant and it was not designed to get actual work done as a coworker. So I started to build Valmis, an alternative to OpenClaw that works with more than 100 apps and services, with security... (38 points, 8 comments).

Kshitij and Rahul here. We're building Playground, an open-source sandbox that lets your non-technical teams prototype directly on production codebases, without bugging you (developers) to spin up and maintain another repo. The repo is here: https://github.com/B1u3B01t/design-playground . We've been working together for six years. Last year we built an AI design agent because we thought designers were the biggest bottleneck. But we were wrong. After talking to teams, we realised designers, PMs,... (16 points, 14 comments).

HF Spaces

16.6K likes

Generate any application by Vibe Coding it DeepSite is a Vibe Coding Platform designed to make coding smarter and more efficient. Tailored for developers, data scientists, and AI engineers, it integrates generative AI into your coding projects to enhance creativity and productivity. DeepSite v4 is a Hugging Face Space tagged with docker, region:us. It has 16617 likes on Hugging Face.

Track, rank and evaluate open LLMs and chatbots Modern React interface for comparing Large Language Models (LLMs) in an open and reproducible way. 📊 Interactive table with advanced sorting and filtering 🔍 Semantic model search 📌 Pin models for comparison 📱 Responsive and modern interface 🎨 Dark/Light mode ⚡️ Optimized performance with virtualization The project is split into two main parts: React Material-UI TanStack Table & Virtual Express.js FastAPI Hugging Face API Docker The application is containerized using Docker and can be run using: Open LLM Leaderboard is a Hugging Face Space tagged with docker, leaderboard, modality:text, submission:automatic, test:public. It has 14038 likes on Hu...

11.2K likes

Create your own AI comic with a single prompt Last release: AI Comic Factory 1.2 The AI Comic Factory has an official website: aicomicfactory.app For more information about my other projects please check linktr.ee/FLNGR. If you like the AI Comic Factory, let me know! I am always creating new spaces and exploring new ideas for demos, meaning I don't have much time to take care of all of them (I wish I could clone myself or ask robots to do it). If you appreciate the AI Comic Factory and would like to leave a tip, that would be very kind 🫶 First, I would like to highlight that everything is open-source (see here, here, here, here). However the project isn't a monolithic Space that can be dupli...

Kolors Virtual Try-On is a Hugging Face Space tagged with gradio, region:us. It has 10125 likes on Hugging Face.

5.1K likes

Wan2.2 Animate is a Hugging Face Space tagged with gradio, region:us. It has 5114 likes on Hugging Face.

5.1K likes

AudioCraft is a PyTorch library for deep learning research on audio generation. AudioCraft contains inference and training code for two state-of-the-art AI generative models producing high-quality audio: AudioGen and MusicGen. Installation AudioCraft requires Python 3.9, PyTorch 2.0.0. To install AudioCraft, you can run the following: We also recommend having ffmpeg installed, either through your system or Anaconda: At the moment, AudioCraft contains the training code and inference code for: MusicGen: A state-of-the-art controllable text-to-music model. AudioGen: A state-of-the-art text-to-sound model. EnCodec: A state-of-the-art high fidelity neural audio codec. Multi Band Diffusion: An EnC...

AI Tools — July 5, 2026 Edition | Agentic Brew