Tools Bench.

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

Last Brew Time: Aug 28, 2026, 11:06 AM PT

Insight

Builders this run are giving coding agents X-ray vision into systems they used to just guess about

01.REVERSE-ENGINEERING HOW SOFTWARE ACTUALLY WORKS

02.VOICE AGENTS GRADUATING INTO PRODUCTION INFRASTRUCTURE

03.A PARALLEL PUSH TO POLICE WHAT AGENTS ARE NOW ALLOWED TO SEE AND DO

Featured

GitHub49.9K

ChromeDevTools/chrome-devtools-mcp

Chrome DevTools for coding agents

youtube
Chrome DevTools MCP with Addy Osmani—GenAI Prompt to Product Showcase

Market Signal

Why It Has Market Pull

Chrome DevTools MCP is an official Google-maintained project that has become a default tool in the AI coding agent ecosystem, giving agents like Claude, Cursor, and Copilot real control over a live Chrome browser for debugging and performance analysis. Adoption is exceptionally strong and verified rather than hyped: nearly 50,000 GitHub stars, endorsement from a senior Chrome engineer, and reported adoption as a default debugging tool at a major enterprise dev team.

  • Roughly 49,900 GitHub stars and 3,500 forks on the official ChromeDevTools org repo
  • Published and promoted directly via the Chrome for Developers blog, giving it first-party Google backing
  • Addy Osmani (Chrome engineering) publicly framed it as turning agents into "loop-closed debuggers"
  • CyberAgent reportedly lists it as their default debugging tool in their agent configuration
  • Listed across multiple MCP registries and directories (glama.ai, npm) with several community forks, showing broad ecosystem pickup

feedbacks

What People Are Saying

  • "turns AI coding assistants from "static suggestion engines into loop-closed debuggers""X post (Addy Osmani)

  • "Coding agents are not able to see what the code they generate actually does when it runs in the browser. They are effectively programming with a blindfold on."Dev.to article

  • "gives your AI coding assistant access to the full power of Chrome DevTools for reliable automation, in-depth debugging, and performance analysis"Chrome for Developers blog

  • "Give Your AI Agent Eyes in the Browser"Substack article

  • "automatically waits for action results using puppeteer, which removes a lot of the flakiness agents used to hit clicking through a page"Dev.to article

  • "lists the DevTools MCP server as their default debugging tool"Dev.to article

GitHub46.1K

abhigyanpatwari/GitNexus

GitNexus: The Zero-Server Code Intelligence Engine - GitNexus is a client-side knowledge graph creator that runs entirely in your browser. Drop in a git repository (Github, Gitlab, Azure, Local) or ZIP file, and get an interactive knowledge graph with a built in Graph RAG Agent. Perfect for code exploration

github
GitNexus — the nervous system for agent context

Market Signal

Why It Has Market Pull

GitNexus is a standout in this set: an open-source, zero-server code-intelligence engine that went from roughly 1,200 to over 45,000 GitHub stars in a matter of weeks, backed by a real Y Combinator (S26) company (Akon Labs) with an active enterprise track — strong evidence of both grassroots developer momentum and a credible business forming around it.

  • 46,000+ GitHub stars and 5,100+ forks, ranked #569 globally on star-history trackers
  • Grew from about 1.2k to 40k+ stars in roughly six weeks, with organic coverage spreading across X, Reddit, and LinkedIn
  • Company behind it, Akon Labs, is a Y Combinator S26 startup offering a managed SaaS/self-hosted enterprise tier with PR review and auto-updating wikis
  • Reported as supported by Anthropic for open source and accepted into YC Startup School
  • Ships 17 MCP tools with native setup for Claude Code, Cursor, Codex, and Antigravity, built on Tree-sitter, an embedded graph database, and Sigma.js visualization

feedbacks

What People Are Saying

  • "everything runs locally on your machine. No network calls"GitHub README

  • "Precomputed structure at index time — clustering, tracing, scoring"GitHub README

  • "License terms for internal use"GitHub issue

  • "gives Claude Code and Cursor full codebase structural awareness"Dev.to article

  • "posts about GitNexus filling up X, Reddit, and LinkedIn feeds, written by people who aren't us"Dev.to article

  • "the nervous system for agent context"Company site

GitHub34.9K

anthropics/claude-plugins-official

Official, Anthropic-managed directory of high quality Claude Code Plugins.

Market Signal

Why It Has Market Pull

This is Anthropic's own first-party plugin directory for Claude Code, giving it about as strong a market-credibility signal as a repository can have; it shows genuine, large-scale developer engagement (tens of thousands of stars, thousands of forks, hundreds of open contributions) even though community sentiment on plugin quality is still mixed.

  • 34,945+ GitHub stars and roughly 3,900 forks
  • 3,438 commits on the main branch with 939 open issues and 110 open pull requests, indicating an actively maintained, high-traffic repository
  • Maintained directly by Anthropic as the official Claude Code plugin marketplace, distinguishing it from third-party alternatives
  • Combines internal Anthropic-built plugins with a vetted external-plugin submission pipeline for partners and the community
  • Hacker News discussion around the Claude Code plugin system notes best practices are still emerging, tempering pure hype

feedbacks

What People Are Saying

  • "35k stars, 3.9k forks"GitHub repo stats

  • "939 open issues, 110 open pull requests"GitHub repo stats

  • "just variations of inserting canned prompts, and no clear best practices have emerged yet"HN comment

  • "/plugin install {plugin-name}@claude-plugins-official"GitHub README

  • "Official, Anthropic-managed directory of high quality Claude Code Plugins"GitHub repo description

  • "3,438 commits on main branch"GitHub repo stats

GitHub13.3K

livekit/agents

A framework for building realtime voice AI agents 🤖🎙️📹

youtube
The End of Annoying AI Interruptions? LiveKit Turn Detector v1 Tested

Market Signal

Why It Has Market Pull

LiveKit Agents is the open-source SDK behind a well-funded, fast-growing voice AI infrastructure company - LiveKit closed a $100M Series C at a $1 billion valuation in January 2026 led by Index Ventures, with Bloomberg noting it sells voice tooling to OpenAI - making it one of the most credible, real-usage-proven products in this space.

  • 13,290 GitHub stars and 3,629 forks, with commits pushed as recently as today and 797 open issues reflecting a large active user base
  • $100M Series C in January 2026 at a $1B valuation, led by Index Ventures with Salesforce Ventures, Altimeter Capital, and Redpoint Ventures participating; ~$174M raised total
  • Bloomberg reports OpenAI is a customer of LiveKit's voice infrastructure, alongside a public LiveKit Cloud platform and Agent Builder product
  • Numerous GitHub issues describe real production deployments (e.g. "running stable in production for weeks" before hitting a worker bug), evidencing genuine operational usage beyond demos
  • Actively shipping features (Expressive mode, open-weights turn-detection model optimized for CPU) rather than a stalled or abandoned project

feedbacks

What People Are Saying

  • "Worker failing over and over again... to be frank I'm irritated now"GitHub issue

  • "An agent was running stable in production for weeks"GitHub issue

  • "audio starts to cut out and eventually disappears"GitHub issue

  • "ease of integration and scalability is helpful"Dev.to article

  • "for production, the more mature Python version of this framework is recommended"Dev.to article

  • "worker hangs indefinitely after logging "process initialized""GitHub issue

GitHub2.5K

tailscale/tailcat

like netcat, but over Tailscale's data plane, without Tailscale's control plane

Market Signal

Why It Has Market Pull

tailcat is a brand-new open-source tool from an established, well-funded company (Tailscale), and it landed a genuinely strong Hacker News debut with substantive technical debate — a credible, worth-tracking release rather than hype, especially for the emerging use case of ephemeral, account-free connectivity for AI agents.

  • Hacker News launch post hit 658 points with 127 comments, drawing engagement from well-known community members
  • Shipped at TailscaleUp 2026 by Brad Fitzpatrick (Tailscale co-founder), who personally addressed vendor-lock-in concerns in the thread
  • 2,500+ GitHub stars within days of release, backed by Tailscale's existing WireGuard/DERP infrastructure and BSD-3-Clause license
  • Explicitly self-hostable without any Tailscale account or control plane, positioned for AI-agent-to-agent connectivity that can't navigate login flows
  • Ships an experimental WebAssembly build for in-browser connectivity and both CLI and Go-library interfaces

feedbacks

What People Are Saying

  • "Interesting. I thought about doing this immediately after reading their old blog post on punching through NAT"HN comment

  • "why Tailscale doesn't go 100% open source to eliminate the ts derp control"HN comment

  • "creating Tailscale-specific tool variants mirrors proprietary software patterns"HN comment

  • "Magic Wormhole but for generalized connectivity, not just file transfer"HN comment

  • "no vendor lock-in and no payment or account required"HN comment

  • "comes with no API or CLI stability promises"GitHub README

Sources

GitHub

DietrichGebert/ponytail

115.1K

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

calesthio/OpenMontage

53.1K

World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your AI coding assistant into a full video production studio.

K-Dense-AI/scientific-agent-skills

36.3K

Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 175,000+ scientists worldwide. 163 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

Product Hunt

Skydive

406

Build cloud agent coworkers that take on real, multi-step work across the tools you already use. Describe the outcome you want, and Skydive creates a working agent in minutes. No code, no prompt engineering, no workflow wiring required. Skydive agents live in your stack, execute repeatable work, and get sharper over time.

Enter Pro

381

Building what runs a business takes more than code. Enter Pro is the AI-native platform that turns an idea into a working product. Plan, build, preview, launch, and scale apps, websites, and custom AI agents in one continuous workspace. Models, databases, authentication, hosting, payments, analytics, and localization are built in—so teams can ship software built for real business, not just prototypes.

Lenz

252

Lenz is an AI fact-checking API for products that cannot afford to hallucinate. It extracts verifiable claims from any text, then checks each one: searching independent sources, running multi-model debate, and routing through a review panel — returning a scored verdict with every source, argument, and step visible. Most AI tools give you one model's best guess from memory. Lenz ensures no single model's blind spots drive the conclusion. Available as API and MCP. Try it free at lenz.io/ph

Speko

247

One API for speech-to-text, language models, and text-to-speech, with public benchmarks beside runtime availability.

Gemini 3.5 Transcribe

224

Our latest speech-to-text model designed for precise and intelligent real-time transcription.

Traccia

192

Traccia is a vendor-neutral AI Agent Control Plane built for teams running autonomous agents in production. Observe agent behavior, evaluate performance, govern actions with policies and runtime controls, and maintain an auditable trail of what happened. Built with an open, developer-first SDK and OpenTelemetry, Traccia works across models, frameworks, and existing observability stacks—so teams can control their agents without being locked into a single AI vendor.

YC Launch

Hacker News

We built open OpenRouter that turns usage into a better model

203 pts

Hi HN, we built an open source model gateway. It's a single place to manage our own self hosted, frontier, and open source models in one place. It’s is rust native, built for concurrency, and implements all the config quirks across models and providers (streaming formats, tool calls, model parameters, rate limits, and different error behavior). The gateway adds under 1 ms for BYOK requests and under 2 ms when Experiential supplies the provider key. It has every major inference provider, and 1000... (203 points, 43 comments).

OzBrain, a shared brain for knowledge between agents and your team

92 pts

I think agent-first chat interfaces will be a primary software modality and busy dashboard/UI will go away. I’m not sure who exactly wins it, but I want my knowledge to grow/go with me. A lot of the “knowledge” ie research, analysis, reasoning will be done by agents as the primary user. Our current notes tools & tasks management systems were built for humans… I don’t care what the 17th thing on my bug backlog is. I want to conduct agents that can execute for me and do great work. What I built Oz... (92 points, 58 comments).

Keenable – A different web search API for AI agents

12 pts

Hey HN! We built https://keenable.ai , a different web search API for AI agents. Keenable searches our own 100B+ page index. We are focused on low cost and latency (p95 <250ms from us-east). We don’t believe in benchmaxxing, so we open-sourced our internal benchmarking suite, NEEDLE (available at https://keenableai.github.io/needle ): a live benchmark that compares Keenable with other search APIs on fresh agent-like queries. I spent seven years at Amazon as a scientist working on web grounding f... (12 points, 5 comments).

Traccia - Observability, Runtime Control & Audit for agents

4 pts

AI applications are becoming agents, which has started to take autonomous decisions. There are plenty of tools and platform available to trace, and observe what an agent or llms calls does. They are good in what they do, but tracing and observability isnt enough for AI agents era. We need a solution that can help you observe, evaluate, create run time policies to govern and finally audit the actions of the agent. We built Traccia to solve this problem. The good part, all of these can be achieved... (4 points, 0 comments).

I shrank DeepSeek V4 Flash to 57GB and it wrote a compiler on my Mac

21 pts

I built a specialized package of DeepSeek V4 Flash 0731 (originally 284B total parameters, 13B active), preserving reasoning, tool calling and coding capabilities: https://huggingface.co/steadfastgaze/DeepSeek-V4-Flash-0731-... I let it write a minimal C compiler targeting ARM64, then test the result with Fibonacci and FizzBuzz programs, and it succeeded in less than 1 hour, with the full recording at: https://youtu.be/XiwSilmV8B0 You can run it on Silicon Macs with my engine https://github.com/... (21 points, 3 comments).

HF Spaces

Qwen-Image-Edit-2511-LoRAs-Fast

2.7K likes

Demo of the Collection of Qwen Image Edit LoRAs Qwen-Image-Edit-2511-LoRAs-Fast is a Hugging Face Space tagged with gradio, mcp-server, region:us. It has 2683 likes on Hugging Face.

Agent Memory Leaderboard

767 likes

Unified memory evaluation · Results expected August 12. Agent Memory Leaderboard · 记忆之巅 A unified, open, and reproducible evaluation platform for long-term memory systems and memory-enabled agents. Agent Memory Leaderboard (AML) compares research methods and commercial products under one evaluation contract. Candidate systems implement memory Add and Search; the official platform fixes Answer, Eval, datasets, models, configurations, result review, and publication. > First public release: The inaugural verified leaderboard is expected to be published on August 12, 2026. > 首期发布： 首期经核验榜单预计将于 2026 年 8 月 12 日发布。 Results are separated along two independent dimensions. Textual and coding tasks use...

Free AI Detector

340 likes

Free AI detector for AI text generation & writing. Lynote. Official Lynote landing page for the free AI detector: paste any text and find out in seconds whether it was written by AI, edited by AI, or written by a human. Get AI-generated, human-written and mixed scores with sentence-level highlights for ChatGPT, GPT-5, Gemini, Claude, DeepSeek and more — free forever, no sign-up, 100% private. This Space is currently a static entry page: Hugging Face now hosts interactive Gradio demos on free CPU only with a PRO subscription, so the working demo (app.py, a dependency-light bilingual statistical detector: burstiness, formulaic phrase density, repeated n-grams, vocabulary uniformity) is kept re...

MiniMax Music 3 Studio

295 likes

MiniMax Music 3 Studio — diffusers demo Streams full songs from lyrics + a structured caption using the MiniMaxMusic3Pipeline diffusers port. The input surface is a single Suno-inspired custom gr.HTML composer (Simple ↔ Studio modes, section-tag chips, structured-caption fields per the official prompting guide) that drives Gradio events via trigger()/props.value; styling uses only theme CSS vars so it follows the Citrus theme natively. Weights: MiniMaxAI/MiniMax-Music3 AoTI kernels: diffusers-internal-dev/MiniMax-Music3-aoti (compiled on RTX Pro 6000, matching ZeroGPU hardware) Generation streams chunk by chunk with a configurable playback headroom. The 8B language-model stage runs eager on....

Microduck Sandbox

114 likes

Real trained RL policies for the Microduck robot, running fully in the browser: MuJoCo compiled to WebAssembly steps the physics, onnxruntime-web runs the policy network at 50 Hz. No server, no backend. Two locomotion variants of the same robot are included: legs (walking, the default) and rollers (the wheeled skating variant). Press M (or hold D-pad up ~1 s on a gamepad) to switch; the roller model, meshes and policies are lazy-loaded on the first switch. | Mode | Checkpoint | What it does | |--------|-----------|--------------| | Run (legs) | BESTalphawalking.onnx | Velocity-tracking locomotion (arrows / WASD to steer) | | Sit | BESTalphasitstand.onnx | Sits down on its hull, stands back u...

Rare Disease, Real Kid: MVA Hackathon 2026

53 likes

Rare disease hackathon 2026 🧬 Rare Disease, Real Kid: The MVA Hackathon 2026 The genome and clinical story here belong to a real child living with Mosaic Variegated Aneuploidy (MVA), an ultra-rare genetic condition affecting fewer than 50 people worldwide. There is currently no established treatment, and care today means managing symptoms. The family has opened the case to the research community, hoping someone can find an answer. Help us understand MVA better! The MVA Hackathon is not intended to provide general medical care, diagnosis, or professional medical advice. Rare Disease, Real Kid: MVA Hackathon 2026 is a Hugging Face Space tagged with gradio, region:us. It has 53 likes on Hugging...

AI Tools — August 28, 2026 Edition | Agentic Brew