WELCOME Happy Tuesday, legends. Welcome back to another edition of The Frontier, our weekly newsletter covering the best new AI launches on Product Hunt. WHAT'S HOT 💸 Stripe is buying the thing that picks your model. Bloomberg reported the OpenRouter deal on Sunday at more than $7 billion, about 5.4x the $1.3 billion the company was worth at its Series B in May. OpenRouter is one endpoint sitting in front of 400+ models, routing each call on price and performance for roughly 8 million developers. Stripe told TechCrunch it doesn't comment on rumors, which is not a denial. ( TechCrunch ) 🩹 An AI agent broke into Snowflake through a bug an AI may have written. Wiz's Red Agent found a script injection in a GitHub Actions workflow on Snowflake's .NET connector, opened an issue with a crafted title, and came out holding a Jira token for engineering and security compliance projects. Wiz says the unsafe change was co-authored by Copilot Autofix. GitHub says that commit touched a different file and the refactor that actually broke it belongs to a named human engineer, which is its own kind of unsettling. ( TNW ) 🐙 Cursor started hosting code on the day GitHub fell over. Origin opened in early beta on Monday for every paid plan: real repositories, pull requests, reviews, branch protections, two-way sync back to GitHub. It landed while GitHub was running 20% error rates on web and API traffic and up to 50% on repository downloads. Cursor's parent Anysphere had closed its $60 billion all-stock sale to SpaceX four days before. ( VentureBeat ) 🏛️ Congress's lawyers now spend longer fixing AI-drafted bills than writing them from scratch. Politico reports the House Office of Legislative Counsel is buried under legislative language that member offices and outside groups generated with Claude and ChatGPT. OLC is itself barred from using AI. The tell showed up in June, when Rep. Anna Paulina Luna's office published an amendment summary with the words "Claude responded" still sitting in it. ( Gizmodo ) 🎙️ ElevenLabs moved voice agent management inside Claude. The new MCP lets you create, list, update, duplicate and delete ElevenLabs agents from the chat rather than the dashboard, and it estimates LLM usage before you commit a change. Sign in to ElevenLabs through Claude and it's there. ( Product Hunt ) Five AI tools you may have missed Taku AI — Borrow the best AI setups and make them yours. 57 • Productivity ▲ 384 Shepherd Terminal — A persistent terminal for Codex and Claude side by side 15 • Developer Tools ▲ 119 Framer AI Agents — Design and publish professional sites with AI 28 • Design Tools Promoted ▲ 892 Atlas by WorkOS — Your AI coworker in Slack 3 • Slack ▲ 110 Open Index — Build Smarter Agents using Structured Context 3 • Open Source ▲ 93 Meterless.ai — Run AI locally and own the whole workflow 5 • Productivity ▲ 83 Ad Your canvas just learned to design Framer 3.0 ships AI Agents that live inside your canvas and can actually design. Drop a brief and watch them build out full pages, create components, write code, connect your CMS, and handle SEO. All without leaving the canvas. They plug straight into Claude Code and Cursor. Yes, inside your actual component system, without breaking it. Try the agents FROM THE FORUMS The fluent ones look guiltiest Ozan Dag (@ozandag) posted his split of bad versus good AI judgment. Bad: avatar job interviews where no human is in the loop, thought leadership that's AI slop in a blazer, emails to people who matter, where the client can always tell, not from the grammar but from the warmth. Good: pulling apart a candidate's portfolio instead of keyword-matching a CV, drafting then rewriting until it's yours, handing it your own strategy and asking it to break it. The room mostly agreed and settled on ownership. Alira Salu (@alira_salu) said the test is whether you'll put your name behind the decision and explain how you got there. Irsa Doham (@irsa_doham) had it tighter: the skill isn't getting AI to answer, it's knowing which questions you shouldn't outsource. Mark Wood (@mark_wood37) added the stakes, because the tool can be perfectly capable and the cost of being slightly wrong still too high. Join the thread

WELCOME Happy Tuesday, legends. Welcome back to another edition of The Frontier, our weekly newsletter covering the best new AI launches on Product Hunt. WHAT'S HOT ⚡ Anthropic will eat your electricity bill. It has formed Theseus Infrastructure with Macquarie and Singapore's GIC to build AI compute sites in the US, the two financiers covering most of the cost per project. The part nobody else has offered: Anthropic absorbs any rise in consumer power prices the sites cause. Community cost backlash is what is stalling data centre approvals across US states, so this is a permitting strategy dressed as a joint venture. ( AI Weekly ) 🪟 Meta put a 30B multimodal model under Apache 2.0. Muse Glimmer is tuned for local agentic tool use, coding and LLM-as-judge work, with 131K context and 100+ languages. Quantised to 4-bit it squeezes under 20GB, which means it runs on one consumer GPU. ( AI Release Tracker ) 🔓 OpenAI is handing out a model with the cyber guardrails off. The expanded Daybreak programme has two tiers: Blue gives you GPT-5.6 Sol with system-level cyber restrictions removed, and Red opens access to GPT-5.6-Cyber, purpose-trained for the work. Offensive security has been the awkward exception to every safety policy, and this is the first time a lab has built the exception its own door. ( LLM Stats ) 🕵️ Someone worked out how to date a frontier model's training run from the outside. A researcher published a method on 10 August that infers training-run identity, placing Anthropic's Opus 4.7+ on a late-December-2025 cutoff and OpenAI's GPT-5.6 family around a late-February-2026 checkpoint. Labs stopped publishing this. Turns out they did not need to. ( LLM Stats ) 🎧 Spotify shipped the tool for running 50 coding agents at once. Xirp manages sessions across Claude Code, Codex and Gemini CLI, each in its own git worktree, with context living in Spotify Portal so you can switch model mid-job and keep the working state. 1,300 Spotify engineers and 36,000 sessions in already, and it is the same play as Backstage, which they open sourced in 2020 and 3,400 companies now run. ( Spotify ) Five AI tools you may have missed Bullet — 30-60% faster than Claude Code and Codex 41 • Productivity ▲ 241 BetterClaw — Deploy AI Agent, 60 seconds & $0 forever 72 • Productivity ▲ 324 Framer AI Agents — Design and publish professional sites with AI 28 • Design Tools Promoted ▲ 892 Gotcha — World's First AI Copilot for Android. You talk. It acts. 3 • Android ▲ 94 Vizard Agent — One AI agent for every kind of video 5 • Productivity ▲ 126 bb — The IDE that builds itself 13 • Developer Tools ▲ 202 Ad Your canvas just learned to design Framer 3.0 ships AI Agents that live inside your canvas and can actually design. Drop a brief and watch them build out full pages, create components, write code, connect your CMS, and handle SEO. All without leaving the canvas. They plug straight into Claude Code and Cursor. Yes, inside your actual component system, without breaking it. Try the agents FROM THE FORUMS The work you would never hand an agent Aleksandar Blazhev (@byalexai) asked what work people would never delegate to an AI agent. The first answers came back as categories: hiring, money, customer relationships, anything creative. Tom McKenzie (@thomastherhombus) called AI the worst brainstorming partner imaginable, and later explained why: it defaults to templated safe output, which is useless when you need a tagline in one brand's voice. Then Gal Dayan (@galdayan), who runs an AI agent company, moved the thread off categories entirely. The line is not what kind of work it is, it is whether anyone downstream can tell the output was wrong before it does damage. Code review is safe to delegate heavily because a bad diff breaks a test or a build and something catches it fast and cheap. A legal filing or a medical read is the opposite, because being wrong looks exactly like being right until it is too late, and there is no build that goes red. Abdullah Javaid (@abdullah_javaid3) took it to content, where a bad paragraph throws no error and just reads fine while saying nothing true. Mio (@iammio) closed it: judgment is really just knowing which of your own outputs to distrust. Good thread if you are deciding right now how much of your pipeline to hand over. Join the thread

WELCOME gm builders, happy Wednesday. This week in four numbers, a prompt worth stealing before Claude changes again, and a forum thread that pulled 94 replies on what actually happens once your agent's mistake reaches someone. THIS WEEK, IN NUMBERS 3 — how many times one agent's "posted" was a lie. The thread → 5 — skill files that just beat 300. The repo → 94 — replies to this week's sharpest AI safety debate. The thread → ½ — what Opus 5 costs against the flagship it nearly matches. The launch → Five builds worth knowing /mission for Claude Code — Give Claude Code missions to spawn a team of agents 40 • Productivity ▲ 292 MemoryCustodian — Repo-native memory for coding agents 28 • Developer Tools ▲ 150 Task Monki — Run coding agents through the full development process 15 • Task Management ▲ 122 Framer AI Agents — Design and publish professional sites with AI 28 • Design Tools Promoted ▲ 892 Prelint — Prevent product drift in AI-written code 135 • Software Engineering ▲ 626 Denovo — Turn your vibe-coded app into paying customers 72 • Sales ▲ 330 STEAL THIS Turn 30 days of Claude Code sessions into a mirror @PrajwalTomar_ built this as a self-audit, not a one-off. Point Claude Code at your own last 30 days of sessions and it turns whatever you kept doing by hand into an actual skill, and flags the biggest inefficiency in how you prompt: Read my last 30 days of Claude Code sessions. Find the top 5 patterns I keep repeating, turn each one into a skill, and tell me the biggest inefficiency in how I prompt. FROM THE FORUMS Your agent does something wrong. How far does it reach before anyone notices? @aadilghani , who builds Pushary, asked builders where they actually draw the line for needing a human in the loop, and got 94 replies in a week. The best ones didn't agree with each other so much as sharpen each other. @raffay_sajjad , building a finance app, realized reversibility was the wrong thing to worry about: you can refund a charge, but you can't un-send the rejection email that went with it. The sharpest reply, from @thenameisarian , reframed the whole question. Reach alone isn't the risk, reach without an accountable name attached to it is. Their fix is a compound gate: if an action's reach crosses a threshold and no named human attached their signature in the last few seconds, it pauses automatically. Not a policy written down somewhere, a mechanism that actually stops the action. Read the thread