Tools Bench.

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

Last Brew Time: Oct 6, 2026, 11:01 AM PT

Insight

The highest real momentum in today's run belongs to single-maintainer Claude Code skills rather than funded startups, as plugin-shaped tooling has quietly become the default packaging format for serious AI developer tools

01.SINGLE-MAINTAINER DESIGN AND WORKFLOW SKILLS

02.PERSISTENT AGENT MEMORY LAYERS

03.LAB-GRADE GPU KERNELS RELEASED AS OPEN REPOS

Featured

GitHub97.0K

thedotmack/claude-mem

Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

youtube
Claude-Mem: Give Claude Code a Permanent Memory

Market Signal

Why It Has Market Pull

This is a real, actively maintained open-source project with extraordinary and verified builder momentum, roughly 97,000 GitHub stars and 8,500+ forks in about 14 months, continuous commits, and multiple independent viral discussions across Hacker News and X. It is well worth tracking closely, though its scale has also surfaced real production risk that any adopter should weigh.

  • 97,089 verified GitHub stars and 8,556 forks as of October 2026
  • Grew from roughly 12,900 stars to 65,800 to 97,000+ across a few months, tracked by multiple independent observers
  • A related Hacker News discussion on agent memory approaches reached 297 points and 173 comments
  • Works across 7+ agent harnesses out of the box: Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode
  • 91 open issues on a project pushing commits same-day

feedbacks

What People Are Saying

  • "Not having to remember to tell Claude to save something is the difference between a tool I use and one I abandon."GitHub discussion

  • "The overall productivity gain of claude-mem is amazing!"GitHub discussion

  • "switch between branches and rebuild knowledge in a few seconds"GitHub discussion

  • "recall quality started degrading"GitHub issue

  • "280 orphaned Claude CLI processes consuming ~65 GB of RAM, ~$183/day in unintended API spend"GitHub issue

  • "instructs an AI coding agent to autonomously read npm credentials and publish without human review"GitHub issue

GitHub77.5K

pbakaus/impeccable

The design language that makes your AI harness better at design.

youtube
Impeccable: El skill que elimina el "AI Slop" de tus interfaces (y es gratis)

Market Signal

Why It Has Market Pull

Impeccable is a genuinely large and active open-source project, built by a credible industry figure, that teaches AI coding agents a shared design vocabulary and deterministic design-quality checks. Its real GitHub star count, active issue tracker, and global ranking confirm strong and sustained builder interest well beyond a launch spike.

  • Verified real GitHub stars: 77,612, with 4,622 forks and commits as recent as October 6, 2026
  • Ranked #231 globally on Star History's tracker
  • Includes 60 deterministic detector rules plus LLM-based critique checks and 24 commands
  • Active issue tracker with real feature requests and bug reports rather than a dormant repo
  • Built by a recognizable ex-Google/Zynga engineering figure, lending credibility

feedbacks

What People Are Saying

  • "love what this tool is doing already, it has so much potential"GitHub discussion

  • "The current installation with npx makes this package a bit unusable. Bring back the zip install"GitHub issue

  • "elements inside an inert subtree under a page overlay count as occluded/nested content"GitHub issue

  • "translucent backgrounds that composite over themselves in the visual contrast walk"GitHub issue

  • ".astro/.jsx/.svelte/.vue scans miss rules that .html catches on byte-identical content"GitHub issue

  • "Impeccable: The Open-Source Design Language With 50,000+ GitHub Stars That Makes AI Coding Agents Better at Design"Dev.to article

GitHub8.6K

deepseek-ai/DeepGEMM

DeepGEMM: clean and efficient BLAS kernel library on GPU

youtube
DEEPSEEK Open Source Week Day THREE - DeepGEMM First Look

Market Signal

Why It Has Market Pull

DeepGEMM is a genuinely high-traction, actively maintained open-source GPU kernel library from DeepSeek AI, a globally significant open-source AI lab. Verified GitHub data confirms real sustained engineering activity, and the library is cited by developers as a reference implementation for clean FP8/BF16 GEMM kernels that power DeepSeek-V3/R1 production inference.

  • 8,659 real GitHub stars and 1,369 forks, pushed as recently as Sept 30, 2026
  • 149 open issues indicating live triage, not an abandoned repo
  • Achieves up to 1,550 FP8 teraFLOPS on Nvidia H800 in roughly 300 lines of core kernel code
  • Released during DeepSeek's Open Source Week and now widely integrated into serving stacks (vLLM, SGLang)
  • Praised by developers as one of the clearest learning references for efficient FP8 kernels relative to CUTLASS

feedbacks

What People Are Saying

  • "clean and efficient FP8 GEMM kernels with fine-grained scaling"GitHub repo description

  • "one of the best learning resources for efficient, clean fp8 kernels"Dev.to article

  • "chose to write a simpler kernel compared to CUTLASS with the same structure and some additional optimizations, with much clearer code"Dev.to article

  • "blew up the AI industry's narrative that more money and power are needed to advance AI"Reddit comment

  • "super impressive"X reply

  • "Independent third-party coverage is still thin on this specific kernel repo beyond developer/technical blogs."evidence gap

Product Hunt158

HyperFrames Studio (Desktop)

The first video editor built for agents - built by the team who open sourced HyperFrames at HeyGen Bring your favorite agent (either codex or claude code), HyperFrames Studio turn your coding agents into video editors, while you stay in the director seat. Describe a video and your agent makes it with HyperFrames, then you and your agent work on it together in one studio editor.

youtube
Claude Video Editing just got a MAJOR upgrade (Hyperframes V2)

Market Signal

Why It Has Market Pull

This is a real product backed by a well-funded company, not a weekend hack. HeyGen, which raised a $60M Series A at a ~$500M valuation, open-sourced the underlying HTML-to-video rendering engine and has now shipped a companion desktop editor that lets a coding agent and a human co-edit the same timeline. The open-source engine already has a large, actively maintained codebase, and the desktop app's launch drew immediate independent press coverage.

  • Underlying open-source engine has roughly 57,900 GitHub stars and 5,100+ forks, with commits as recent as October 6, 2026
  • Desktop app launched on Product Hunt October 5, 2026 and collected 158 votes
  • Parent company HeyGen raised a $60M Series A at a $500M valuation from Benchmark, Thrive Capital and Bond
  • Independent coverage within 24 hours from multiple technical blogs and newsletters
  • 50+ open issues on the engine repo show ongoing day-to-day usage

feedbacks

What People Are Saying

  • "Fatal JavaScript invalid size error on large compiled HTML since version 0.8.77"GitHub issue

  • "Describe it, your agent builds, then you both work on the same timeline."X post

  • "Video as Code: A Deep Dive into HeyGen's Hyperframes"Dev.to article

  • "AI Agent Makes Unlimited Free Videos (Hyperframes Tutorial)"Dev.to article

  • "HeyGen's HyperFrames adds a shared timeline for AI-made video"Press article

  • "Hyperframes vs Remotion"GitHub issue

  • "Improve quickstart documentation"GitHub issue

Sources

GitHub

mattpocock/skills

277.9K

Skills for Real Engineers. Straight from my .agents directory.

msitarzewski/agency-agents

157.7K

A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.

ayghri/i-have-adhd

54.3K

A skill to stop your coding agent from burying the answer. ADHD-friendly output.

cathrynlavery/diagram-design

43.8K

Editorial diagram design for Claude Code, Codex, GitHub Copilot, Factory Droid, and Pi. 42 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.

tester-army/e2e

6.0K

Next generation e2e testing framework for web and mobile apps.

Product Hunt

Spira Maxima

436

Spira Maxima turns plain text into a finished social video: presenter, on-trend B-roll, captions and music in one pass, ready to post with nothing to re-edit. Bring a photo and voice for your AI clone, product shots, or a recording, and it builds the edit around you. Post-trained on social trend and performance data, so every video follows what is working to maximize your social impact.

FastRouter.ai

354

FastRouter is a unified AI gateway and control plane for developers and enterprise teams building with LLMs. It routes every request to the right model across 200+ LLMs through one OpenAI-compatible API, optimizing for cost, latency, quality, and reliability. With intelligent routing, failover, observability, and governance, teams can scale AI apps without vendor lock-in or code changes.

Invofox Self Serve

336

Invofox turns any document into clean, structured JSON through one API. Get 99%+ accuracy backed by SLAs. If we make a mistake, you don’t pay for that document. Invofox handles parsing, extraction, validation, edge cases, and monitoring behind one endpoint, then automatically learns from your feedback to keep improving on the documents your business actually processes.

Opengeni

325

Ship AI agents to production in hours. Opengeni gives you sessions that recover from failures, isolated sandboxes, 100+ integrations, human approvals, and visibility into every step and dollar spent. Focus on your agents. We handle the infrastructure. Self-host for free, or get started on our cloud. For this launch, the first 100 users get $100 in cloud credit. Promo-code: PRODUCTHUNT100

Web Search API

169

Web Search API lets your AI agents and applications search the Internet and ground their responses in live information, instead of guessing URLs or relying on a model's training cutoff.

Siteprint

107

Siteprint is a Safari extension for Mac that measures the design of any page: colours with their roles, type scale, spacing, corners and layout. Extract it as one exact prompt for Claude Code, Cursor or Codex. Runs on your Mac. $9.99 once.

YC Launch

Roma: The to-do list that does itself

4

We're reinventing the to-do list so it does the work for you. Roma · Fall 2026 · Consumer Tags: Consumer, Productivity, AI Assistant. Website: https://roma.app

Linc: The discovery layer for enterprise AI

3

Linc is an AI-native process intelligence platform that captures workflows and unwritten business rules, identifies where AI can create value, and provides the context to redesign processes and build agents. Linc. · Summer 2023 · B2B Website: https://withlinc.com

Rhem Labs: Health assistant robots for your parents

2

Rhem handles everything that doesn’t need a pair of hands, while keeping you in the loop. Rhem Labs · Fall 2026 · Healthcare Tags: Robotics, Health & Wellness, Consumer Products. Website: https://rhem.ai

Hacker News

HN.watch – Videos of all Hacker News posts

225 pts

Hi HN, I’m Per, founder of Scrimba (YC S20). We’ve spent the last decade teaching people how to code with an HTML-based video format. We’ve now plugged an LLM into it, so that people can create explainer videos about anything. It’s called “Scrimba Explain”. To demo this technology for Hacker News, we built HN.watch. It’s like HN, but with explainer videos instead of articles. We create them on-the-fly the first time someone clicks on a link. While there are obvious visual drawbacks of using HTML... (225 points, 97 comments).

OpenChart – OSS TradingView alternative with your own AI agent

29 pts

Hi HN, I'm Weilun, cofounder of OpenChart. We built OpenChart because we wanted Claude and Codex to interface with the markets, like most of humans do, through charts rather than through CLIs. With your existing AI plans, your agent can work directly with your charts: annotate a setup, write an indicator, investigate a market move, or create an alert. The chart, conversation, and research live in the same workspace. OpenChart alerts on any market move: draw any shape on a chart and attach an ale... (29 points, 4 comments).

OpenAPPA – open-source deterministic guardrails that don't break agents

25 pts

Hi Hacker News! Matvey, one of the authors, is here. While building enterprise agents, we ran into a problem: the more tools you connect to the AI, the higher the chance it will run out of control and leak sensitive data. Guardrails, in theory, should prevent this, but the situation is worrying: - Non-deterministic guardrails (LLM as a judge, auto modes, etc.) are vulnerable to prompt injections, or they lack knowledge of the data, making them inefficient (~10% data leaks on our benchmarks). - E... (25 points, 12 comments).

PlaceCall (YC W26) – agentic API to call businesses and get things done

16 pts

Hi HN, this is Yarik and Vlad from VOYGR - we are building the tools for agents and apps to engage with local businesses. It all started with our own pain point at VOYGR: calling businesses to verify if they are open. We are both from Google (Maps and Search) and even there, the merchants and venues don’t keep this info updated. So we built an API and started using it in-house. On July 4th, we were driving through Portland looking for a place to eat. Google Maps was saying “Holiday hours may var... (16 points, 4 comments).

MCPJam - the first testing & evaluations platform for MCP servers

13 pts

Prathmesh, CEO of MCPJam here. Users now start in ChatGPT, Claude, Cursor, and other AI clients. They reach your product through your MCP server. That means your users often aren’t in your product anymore. You can’t see what they prompted for, how the agent interpreted it, or whether your server helped them get the result they wanted. I saw this firsthand leading MCP technical strategy at Asana, including our ChatGPT and Claude launches. We were building high-stakes enterprise integrations, but... (13 points, 9 comments).

Praxos – Multiplayer AI

8 pts

Hey Hacker News! Lucas here, founder of Praxos (YC S24). Praxos is a team messaging platform for people and AI agents. It offers people and AI agents a place to talk and work together via a messaging platform that remembers the context around conversations. That context can then be used by the next person, AI agent, or even you, a week later. You can pick work back up without needing to get hold of another person to explain things again for you… or give you a refresher. A surprising amount of wo... (8 points, 0 comments).

HF Spaces

Jev Decision Index

465 likes

Benchmarks and news on various repros of TypeSafe's Jev Who is rebuilding TypeSafe's Jev (System One / RLCD) in the open? This static Space opens on the Decision Index leaderboard; the News tab tracks the artifacts in one combined grid, color-coded by kind: Decoding: parallel constrained decoding on stock models (inference technique, no new weights) Diffusion: text diffusion models run in a "Jev mode" Trained: Jev-like scoring heads and fine-tunes, weights often on the Hub, promised models listed last Prior art: "this already exists" claims Explainers: architecture speculation, explainers, benchmarks and roundups Cards sort by a trending score: ♥ likes on X + 5 × GitHub stars + 8 × Hub likes...

Qwen-Image-2.1

386 likes

Interactive demo for Qwen-Image-2.1 — unified text-to-image generation and image editing with native RGBA transparency support. 📑 Blog 🤗 Model Weights 💻 GitHub Qwen-Image-2.1 is a Hugging Face Space tagged with gradio, region:us. It has 386 likes on Hugging Face.

OpenVuln

209 likes

Find bugs in your repository with GLM This Space is built automatically from the root Dockerfile and serves the Vite application with nginx on port 7860. Optional Space build variables: VITEAPIBASEURL — API origin; defaults to https://openvuln.vulnhunter.pro. VITEGITHUBREPOURL — source repository linked from the interface. OpenVuln is a Hugging Face Space tagged with docker, region:us. It has 209 likes on Hugging Face.

Viggle Turbo for Qwen-Image-2.1

202 likes

6-step Qwen-Image-2.1, T2I + editing, vs-base comparison Viggle Turbo v0.3 — 6-step Qwen-Image-2.1 A distilled Qwen-Image-2.1 that generates and edits images in 6 steps with no classifier-free guidance, about 5× faster than the 40-step base model. On most prompts it is hard to tell apart from the base model; small, dense text and complicated edits (multi-reference composition, face swaps, identity-preserving edits) can still fall short of it. v0.3 (2026-09-29): at 6 steps, less grain than v0.2.1 and a little softer on fine texture. We think 6 steps is close to its capacity: every further gain we found cost something elsewhere. The new 9-step setting runs 7 turbo steps and lets the base model...

JEV-27B Live Demo

94 likes

Play Mario, Rubik's Cube and Tetris with JEV-27B Launch a live JEV-27B game run in the game arena. Mario: original NES World 1-1, with movement and jump decisions. 3D Rubik's Cube: a 25-turn scramble; select a seed or create a new scramble. Tetris: smooth gravity acceleration, continuing at maximum speed until 20 lines. The right panel shows the selected action, option probabilities, and current / average individual model inference duration. Mario movement and jump are separate calls. Start and stop runs yourself; one run per game executes at a time, with a short queue. Recent runs reconnect when you reload the page. The game worker runs on AutoTrust's existing B300 deployment. Model inputs...

MiniMax H3 Turbo LoRA

1.2K likes

Video generation with a synchronized soundtrack MiniMax-H3 — unquantized, split across two Spaces Joint video and soundtrack out of a single denoising pass, at bfloat16 with no quantization anywhere. This Space is the denoising half: the 61.73 GiB transformer and the two autoencoders. The 62.14 GiB Qwen3-VL conditioner runs in qwen3vl-conditioner, which this Space calls over the gradio API for every request. The weights are the public MiniMaxAI/MiniMax-H3 diffusers checkpoint. MiniMax-H3 is 195.9 GiB in bfloat16 and a ZeroGPU Space is evicted at 150 GB of storage. An unquantized single Space is therefore impossible, which is why quantized demos of it run NVFP4 or float8 weights. Cut the Mini...