SITE MAP
Every page,
in one place.
19 page types and 647 pages.
The Brief (front page)
The day in 3, community picks, latest news, the directory, checks and Radar
AI Tools
278 tools, search, rating filter, grouped by category
Categories index
17 categories + the agents comparison
Category page
17 pages, for example Frontier lab / LLM.
All 17
- Frontier lab / LLM
- AI assistant
- AI agent
- App builder / no-code
- Coding agent / dev tool
- Video & avatars
- Image & design
- Voice & audio
- Automation & integrations
- Data & scraping
- Research & analytics
- Marketing & sales
- Dev infrastructure / APIs
- Agent infrastructure
- Productivity & knowledge
- Education & community
- Fintech / crypto
Tool page
278 pages, for example 1Password.
All 278
- 1Password
- Adobe Firefly
- AgentMail
- AI SDK (Vercel)
- Artificial Analysis
- Base44
- Boardy
- bolt.new
- Canva
- ChatGPT (OpenAI)
- Claude & Claude Code (Anthropic)
- Claude Duet (Anthropic)
- Clay
- Composio
- Cursor
- ElevenLabs
- fal
- Firecrawl
- Gamma
- GitHub
- GitHub Copilot
- Google AI Studio
- Google Antigravity
- Google Cloud
- Google Gemini (app)
- Granola
- Grok / SpaceXAI (formerly xAI)
- Grok Bot
- HeyGen
- HubSpot
- Intercom (Fin AI Agent)
- Krea
- Lindy
- LiveKit
- Lovable
- Magnific (formerly Freepik)
- Make
- Manus
- Microsoft Copilot
- Mirage (Captions)
- Mistral AI
- n8n
- Notion
- OpenAI Developers (OpenAI Platform & Codex)
- OpenRouter
- Perplexity Computer
- Replit
- Runway
- Stitch by Google
- Supabase
- Vercel
- Zapier
- Agent Relay
- agent-desktop
- Agentcard (opencommerce)
- Ahrefs
- Arcads
- Attio
- Baseten
- Black Forest Labs (FLUX)
- Blotato
- Blume
- Browser Use
- Browserbase
- Buzz (Block)
- Cap
- CapCut
- Cartesia
- Cekura
- changedetection.io
- Chrome built-in AI (Gemini Nano)
- Clean (Clean AI Labs)
- CleanShot
- CLI-Anything
- Cloro
- Cloudflare (Workers AI)
- Cohere
- Coldtea
- ComfyUI / Comfy
- Context.dev
- Convex
- CopilotKit
- Crawl4AI
- CrowdReply
- Deepgram
- DeepSeek
- Descript
- Dessa (Block)
- Devin / DeepWiki (Cognition)
- Devpost
- ego(lite)
- Elu
- Emergent
- Exa
- Factory (Droid)
- Fastlane
- Figma
- Fish Audio
- FocuSee
- Gemini API CLI (Google)
- Gemini Notebook (formerly NotebookLM)
- Gemma (Google DeepMind)
- Genie (Google DeepMind)
- Genspark
- Glif
- Gojiberry
- Google ADK (Agent Development Kit)
- Google AI Edge Foresight
- Google Flow
- gpt-oss (OpenAI open-weight models)
- Gradium
- Harness2 (Google Research)
- HarnessRouter
- Higgsfield AI
- Hugging Face
- Inbox Zero
- Inkbox
- InsForge
- Instantly
- Julius AI
- Kilo Code
- Kimi (Moonshot AI)
- Kling AI
- Knock
- Kylon
- Lightreel AI
- Liquid AI
- Listen Labs
- Lovart
- Luma
- MagicPath
- MakeUGC
- Manufact (mcp-use)
- Marketing Skills (Corey Haines)
- MaxFusion
- Mem0
- Meta AI (Llama)
- Meta Muse
- Midjourney
- MiniMax
- Mintive
- Modal
- Monid
- MonoDesk (Envato)
- Mosaic
- Nous Research
- NVIDIA Nemotron 3.5 Transcribe
- Obsidian
- Ollama
- OpenAI Agents SDK (TypeScript)
- OpenClaw
- OpenCode
- OpenShip
- Optimizely
- OpusClip
- Ora
- Osaurus
- Otter.ai
- Pamba
- Perplexity Comet
- Photon
- Pika
- Ploid
- Poke
- Polar
- Polsia
- Profound
- Prompt Motion
- Qwen (Alibaba)
- React Bits
- Recursive (Flint)
- Reflection
- Remotion
- Requesty
- Resend
- Result
- Riverside Agents (Riverside)
- Savee
- Screen Studio
- Screenpipe
- Seedance (ByteDance)
- Sentient OS
- Shipper
- SideShift
- SkillSpector (NVIDIA)
- Snaptrude
- Speechmatics
- Suno
- Sunra
- Superdesign
- Superhuman (Grammarly suite)
- Synthesia
- Tables
- Tavus
- TesterArmy (e2e)
- Timestral (Mistral AI)
- TinyFish
- Together AI
- Town
- tstack (Theo Browne's agent skills)
- TypeSafe AI (Jev)
- Unify
- Vapi
- VEED
- WanGP (Wan2GP)
- WebMCP (Google Chrome)
- Wispr Flow
- 1Claw
- Agent Native
- agent-browser
- Agentbox
- Applied Compute
- AQuA – Ambient Quality Agent (Google ADK)
- Arena
- Awesome LLM Apps
- Axiom
- Blaxel
- Boring Funnels
- Buildcamp
- Capafy
- CodedVisuals (pixelcave)
- CodeWords
- Concurrence
- Content Rewards
- Desert Ant Labs
- Despia
- DigitalOcean
- Dreambase
- Droxy
- Flue
- FriendliAI
- fx0 (Vercel Labs)
- Goshen Email
- GrowFi
- Harvey
- Hi3D
- Hy-Memory (Tencent Hunyuan)
- Island Enterprise Browser for Agents
- json-render
- Junie (JetBrains)
- meaningtech (Fluxkit, Growup)
- Memo
- Moca Network
- Mubit
- Nanbeige (Nanbeige LLM Lab)
- Nectar
- nekuda
- Novita AI
- outbid.lol
- Overmind
- Percolator
- Pexo
- PhoneUse
- portless
- PraisonAI
- Real Vision AI Trade Agent
- Repo Prompt
- Robyn
- ScaleBot
- scriptc (Vercel Labs)
- Sequential Thinking MCP
- Serro
- shadcncraft
- Shaders
- shim-cli
- snapcn (Shadcn Labs)
- supastarter
- supatools.dev
- Sutyo AI
- Thinking Machines Lab
- Treg
- UI Rules
- Visko
- vLLM
- Vyb (Coco Labs)
- wafer
- Bud (formerly Orchids)
Personal AI agents comparison
20 agents, filters and quick picks
Latest news
Live news briefings, topic filters and search
Community picks
80 collected stories
News archive
11 earlier editions
News story
251 pages, for example 1 Nov: ChatGPT and Codex will share one usage allowance from December.
All 251
- 1 Nov: ChatGPT and Codex will share one usage allowance from December
- 1 Nov: About 1 in 17 public agent skills tells the agent not to ask you first
- 1 Nov: Anthropic: Claude agents learned to sneak bad training data past a filter
- 1 Nov: Don't believe it yet: the 'Claude Fable 5 weights leak'
- 1 Nov: Grok Bot for LinkedIn: hype or fact?
- 1 Nov: Infostealer malware now takes AI coding-tool keys
- 1 Nov: Suno MCP: make music from inside Claude or ChatGPT (beta)
- 1 Nov: Island launches a company-controlled browser for AI agents
- 1 Nov: Do you need to build your own agent harness?
- 1 Nov: Factory's Droid 1.0: one program instead of hundreds of tool calls
- 1 Nov: Watch out: recycled 'free course' videos
- 1 Nov: The cobra effect: what metrics teach us about managing AI agents
- 31 Oct: Google DeepMind's PlanAssist heads to 15 English councils to speed up planning
- 31 Oct: Vercel Labs releases agent-desktop, a free way for AI agents to use desktop apps
- 31 Oct: Claude's plans get new names, and Ultra loses the 5-hour limit
- 31 Oct: A tampered Claude Code extension reached Cursor's marketplace
- 31 Oct: OpenAI Private Intelligence: tighter data controls for businesses
- 31 Oct: Gemini Notebook adds free textbooks as sources
- 31 Oct: Agentcard opencommerce: one 'buy' tool for agents
- 31 Oct: Glif: consistent AI models for your ads
- 31 Oct: @kimmonismus: Anthropic's week, Reddit versus AI Mode, and gadgets
- 31 Oct: @steipete: own your harness
- 31 Oct: @mattpocockuk: an agent building him a watch app
- 31 Oct: @Saboo_Shubham_: what an agent harness actually is
- 31 Oct: @Sabrina_Ramonov: a prompt for tailoring a CV
- 31 Oct: @mikefutia: a walkthrough of ScaleBot's Meta ads analytics
- 31 Oct: @dr_cintas: seven free open-source tools
- 31 Oct: @gregisenberg: the future of SaaS in one image
- 30 Oct: OpenAI doubles limits on its top plans and drops the Pro 200 cut
- 30 Oct: Mistral Large 4 ('Le Chonk') is out as open weights
- 30 Oct: Granola makes voice typing free in any Mac app
- 30 Oct: Ploid: people search for agents, with a big UK privacy question
- 30 Oct: Buzz by Block: a team chat where agents are members
- 30 Oct: Osaurus plus LFM3: a private AI that runs on your Mac with Wi-Fi off
- 30 Oct: Gojiberry Sales OS: a sales team written as skills
- 30 Oct: @kimmonismus: 'We won', the confusing maths, and some rumours
- 30 Oct: @Mr_Salio: a possible shared ChatGPT and Codex limit
- 30 Oct: @dr_cintas: Robyn, Serro, Snaptrude and Pika
- 30 Oct: @ctatedev: Vercel Labs CLIs can now act as MCP tools
- 30 Oct: @skirano: an agent-collaboration prototype built overnight with Codex
- 29 Oct: Google announces Gemini 4 Carbon, but you can't use it yet
- 29 Oct: Free API credits if you pay for ChatGPT Pro or Business
- 29 Oct: Claude for Legal gets a gallery of ready-made tools
- 29 Oct: GitHub Copilot Fusion: pick an effort level, not a model
- 29 Oct: NVIDIA's open transcription model labels who said what
- 29 Oct: Mistral launches a Responses API that mirrors OpenAI's
- 29 Oct: Quick lab notes
- 29 Oct: Kylon: shared rooms where your team and your agents work together
- 29 Oct: Unify: the agent prepares the call, a person makes it
- 29 Oct: A near miss worth sharing: never auto-approve an agent in someone else's repo
- 29 Oct: Supabase Compute: run agents next to your database
- 29 Oct: Context.dev Parsers: paste a Google Maps, TikTok or G2 link, get clean data
- 29 Oct: Matt Pocock: make good code the only path for your agent
- 29 Oct: Vercel CI: build checks that agents can set up and read
- 29 Oct: Mike Futia: from Shopify reviews to Meta ad angles, inside ScaleBot
- 29 Oct: Claude for Legal: a regulatory horizon scan in a few clicks
- 29 Oct: Write an honest 'alternatives to [competitor]' page
- 29 Oct: @kimmonismus: benchmark context for Carbon's 'number one'
- 29 Oct: @Mr_Salio: Carbon pricing and the Pro 200 allowance cut
- 29 Oct: @bcherny: Utah's Claude Code pilot
- 29 Oct: @dr_cintas: Gamma's new agent makes an Instagram carousel, and VEED's dubbing API
- 29 Oct: @Saboo_Shubham_: Google's token numbers
- 29 Oct: @mikefutia: review scraping for ad ideas
- 28 Oct: Google DeepMind's Genie 4: a video you can walk into
- 28 Oct: Anthropic moves into drug discovery, and buys a one-person biotech run with Claude
- 28 Oct: Gemini API CLI: Google's media models in one terminal command
- 28 Oct: ChatGPT can now read your iCloud Mail and Calendar
- 28 Oct: Anthropic: Claude at 40% off for US government
- 28 Oct: Now Anthropic claims an AI refuted a major computer-science hypothesis too
- 28 Oct: Quick lab notes
- 28 Oct: WebMCP isn't replacing MCP: use both
- 28 Oct: Sentient OS: a private, on-device 'second brain' for Macs
- 28 Oct: Concurrence: agents that take patient histories before appointments
- 28 Oct: Axiom's agents formally checked one of maths' biggest proofs in 24 days
- 28 Oct: Supabase scoped access tokens: give your AI agent one project, not your whole account
- 28 Oct: Cursor's design mode gets Annotate: draw on the page, the agent fixes it
- 28 Oct: Requesty Auto Routing: let a router pick the right model for each Claude Code session
- 28 Oct: Matt Pocock: what makes a task good for an agent
- 28 Oct: Mike Futia's ScaleBot Ad Spy: reading competitors' Meta ads
- 28 Oct: Andrej Karpathy at Sequoia's AI Ascent: how to work with agents
- 28 Oct: A Stanford lecture on 'more attempts plus a good checker'
- 28 Oct: Find the AI traffic you already get in Google Analytics
- 28 Oct: @gregisenberg: 'SaaS is becoming free'
- 28 Oct: @kimmonismus: drug discovery as Anthropic's next big revenue stream, and Dario on leaving OpenAI
- 28 Oct: @Mr_Salio and @skirano: GPT-6.1 Terra usage limits
- 28 Oct: @dr_cintas: Dreamina 2.0's node canvas
- 28 Oct: @mikefutia: a one-shot, fully AI Meta ad
- 28 Oct: @steipete and @ctatedev: two quick developer notes
- 28 Oct: @Saboo_Shubham_: Gemini 4 Argon vs GPT-6 Astra on research tasks
- 27 Oct: Anthropic files to go public, with its first operating profit (not a net profit)
- 27 Oct: OpenAI's GPT-6.1 Terra: a cheaper, faster model that can use your screen
- 27 Oct: Cohere catches an open model trying to 'phone home' from a sandbox
- 27 Oct: OpenAI says its models produced new results in maths, computer science and physics
- 27 Oct: Rumour roundup (none of it confirmed)
- 27 Oct: Cursor's Agent SDK is now generally available: agents as code, with tests
- 27 Oct: BELL: Browserbase's open-source blueprint for a personal assistant
- 27 Oct: Theo's tstack: nine open-source skills for Claude Code and Codex
- 27 Oct: NVIDIA SkillSpector: scan skills before you install them
- 27 Oct: Claude for Nonprofits: a ready-made skill set for one sector
- 27 Oct: Riverside Agents: a team of AI agents for editing your recordings
- 27 Oct: Hi3D V1.0: a free, open image-to-3D model
- 27 Oct: Matt Pocock's prompt of the day: turn a one-way door into a two-way door
- 27 Oct: Greg Isenberg and 'Boring Marketer': how to sell AI to Main Street businesses
- 27 Oct: Theo's tstack: the agent skills he uses to review and merge code
- 27 Oct: Google's free Graph RAG agent codelab (reposted as a 'course')
- 27 Oct: Win the specific question, not 'best [category]'
- 27 Oct: @gregisenberg: his takeaways from Anthropic's S-1
- 27 Oct: @thsottiaux, @kimmonismus and @Mr_Salio: is GPT-6.1 Terra any good?
- 27 Oct: @Mr_Salio: Opus 5.5 builds a 3D world in 10 hours
- 26 Oct: Kimi says sorry for K3.1 Pro rate limits, but not for the influencer brief
- 26 Oct: CapCut is coming to ChatGPT (and part of it already works, if you're on the right setup)
- 26 Oct: Nanbeige4.3-33B: a free open model aimed at office and coding agents
- 26 Oct: Rumour roundup: what might land this week (none of it confirmed)
- 26 Oct: Don't wait for your agents; let them wait for you
- 26 Oct: People are using Claude Code to strip ads and bloatware off their smart TVs
- 26 Oct: Matt Pocock: a 'software factory' running on GitHub Actions
- 26 Oct: Jake Dabrowski's Claude Code workflow for product videos (dropshipping)
- 26 Oct: A real Stanford lecture on LLMs and agents, wrapped in a fake '$100k agent' post
- 26 Oct: The 'ready-made Claude Code teams' clip: what it really shows
- 26 Oct: Answer the follow-up questions, because Google's AI searches for them too
- 26 Oct: @nntaleb: AI will show up in hiring before it shows up in layoffs
- 26 Oct: @gregisenberg: 'Sent from AI agent' is the new 'Sent from iPhone'
- 25 Oct: Kimi K3.1 Pro: a leaked brief alleges a paid, unlabelled influencer campaign
- 25 Oct: OpenAI is using discounts to win Anthropic's coding customers
- 25 Oct: How Anthropic keeps Claude Duet away from your passwords and money
- 25 Oct: OpenAI's Codex CLI 0.164 adds reusable prompt presets
- 25 Oct: Rumours from the leak trackers: Gemini 4 'Carbon', a new Google image model and Qwen 4
- 25 Oct: Report: Perplexity's 'Taste' skill is free until 31 October
- 25 Oct: HarnessRouter: one gateway to run Claude Code, Codex and other agents
- 25 Oct: CLI-Anything: let coding agents drive your desktop apps
- 25 Oct: Sutyo AI: a 'personal memory' app
- 25 Oct: Harness engineering paper: the 7 building blocks every coding agent shares
- 25 Oct: Mariah Brunner's AI motion graphics starter kit
- 25 Oct: Lee Robinson's Stanford lecture: how always-on AI agents work
- 25 Oct: Google's Agent102 lab: build a self-improving agent (not the '2-hour course' the post promises)
- 25 Oct: HarnessRouter demo: one agent hands a data job to two specialist agents
- 25 Oct: Ask each AI assistant who it recommends in your area: they rarely agree
- 25 Oct: @Sabrina_Ramonov: a mock-interview prompt and 10 study commands for ChatGPT
- 25 Oct: @mikefutia: an AI b-roll generator inside ScaleBot
- 24 Oct: Kimi K3.1 Pro: a huge Chinese model at about half Claude Opus prices
- 24 Oct: The labs are fencing in their own AI: OpenAI cut the internet from training, Anthropic details a misbehaving model
- 24 Oct: Meanwhile, everyday agents get longer leashes: Claude MCP Tasks, Cursor computer use and Agent Relay
- 24 Oct: Block's Dessa: a voice AI that listens while it talks
- 24 Oct: OpenAI resets Codex limits again as Uber goes all in
- 24 Oct: Report: Anthropic may be testing different Claude Max limits
- 24 Oct: Three lessons from rebuilding computers for AI agents
- 24 Oct: Security roundup: a medical lab breach and a warning from Cloudflare's CEO
- 24 Oct: Report: OpenAI engineers replacing custom code with Codex skills
- 24 Oct: macOS 27 has a free AI model in the Terminal
- 24 Oct: Mubit: small custom models for the jobs your agents keep repeating
- 24 Oct: Harness engineering: the term to know this autumn
- 24 Oct: Gemini 4 Argon builds a 3D cathedral from one image
- 24 Oct: Mike Futia: a 'Yapper' talking-head UGC ad made with Seedance 2.5
- 24 Oct: Anthropic's Lydia Hallie on cutting Claude Code costs (the post's '65 agents in a loop' claim is false)
- 24 Oct: Cursor Cloud Agents webinar with Growth Engine X (don't trust the 'SpaceX AI engineer' framing)
- 24 Oct: Make your pages easy for AI to quote, and show a real person keeps them up to date
- 24 Oct: @gregisenberg: the most obvious startup ideas right now
- 24 Oct: @Sabrina_Ramonov: make ChatGPT write like you by default
- 23 Oct: Anthropic hands Agent Skills to the Linux Foundation, so your skills travel between tools
- 23 Oct: Big US banks are finding unvetted AI agents inside customer accounts
- 23 Oct: Claude's apps switch to 'auto mode' by default from next week
- 23 Oct: Claude is now generally available on Google Cloud, run by Anthropic
- 23 Oct: Google: a Gemini command-line tool for agents, and a computer for every agent task
- 23 Oct: Anthropic used Claude to prove security properties in 100+ open-source libraries
- 23 Oct: OpenAI's Decisions API is now generally available
- 23 Oct: Independent testers found ways OpenAI's internal agents could slip past monitoring
- 23 Oct: An Anthropic security leader warns that AI-powered attacks will get worse before better
- 23 Oct: Matt Pocock's /grill-me skill: let the agent interview you first
- 23 Oct: Supabase: give your app its own MCP server
- 23 Oct: 1Password for AI agents: approve logins without handing over passwords
- 23 Oct: Canva Craft: Canva's first in-house image model
- 23 Oct: MonoDesk: Envato's all-in-one workspace for freelancers and small studios
- 23 Oct: PhoneUse: a real phone for your AI agent
- 23 Oct: Teleport's five security-review prompts (sponsored)
- 23 Oct: Eric Siu's open-source '10-expert panel' prompt for grading SEO plans
- 23 Oct: From Gemini app to a live website with Google Sites
- 23 Oct: Tech With Tim: Claude Code builds a job-market tracker from 40,000+ Hacker News posts (sponsored by Databricks)
- 23 Oct: Anthropic's Jess Yan on building long-running agents (a manipulated re-upload: watch Peter Yang's original instead)
- 23 Oct: Your About page may be your most-cited page in AI answers
- 23 Oct: @kimmonismus: AI that listens while it talks
- 23 Oct: @nntaleb: a stark warning on white-collar jobs
- 22 Oct: Claude Duet: Anthropic's proactive agent app (US waitlist only, not in the UK yet)
- 22 Oct: OpenAI says its models helped find a critical PHP flaw, and there's a Codex skill to check your site
- 22 Oct: Recursive's Flint: a free coding model that runs offline on your laptop
- 22 Oct: Claude Security now re-scans your code every time it changes
- 22 Oct: OpenAI's models are coming to Google Cloud
- 22 Oct: Gemini 4 Argon builds a 3D anatomy explorer, plus a Gemini 4 Carbon rumour
- 22 Oct: NVIDIA at GTC Berlin: helping agents know your business
- 22 Oct: Railway security incident: if you host apps or agents there, rotate your keys
- 22 Oct: Check what your agent skills could leak
- 22 Oct: Capafy: a marketplace for selling agent skills as products
- 22 Oct: ElevenCreative Chimera: talking-character videos with expressive AI voices
- 22 Oct: Firecrawl People Data: free prospect research inside ChatGPT and Claude
- 22 Oct: Google's Harness²: an agent that improves its own set-up for each task
- 22 Oct: Greg Isenberg: getting your business ready for AI shopping agents (31-minute strategy talk)
- 22 Oct: Nidhi Singh: let Claude Code use Codex's computer use to drive Mac apps (includes a sponsored segment)
- 22 Oct: Claude Code + Obsidian: keep your agent's memory in a notes vault (likely a re-upload; credit @0xrux)
- 22 Oct: Ask your readers to make you a Google 'Preferred Source'
- 22 Oct: @dr_cintas: ten starter jobs for Claude Duet
- 22 Oct: @thsottiaux: Codex and Work now keep your prompt cache permanently
- 22 Oct: @mattpocockuk: three steps to becoming a real software builder
- 22 Oct: @kimmonismus: AI agents reportedly used in attacks on 400+ Latin American banks
- 22 Oct: @kimmonismus (sponsored): Junie Lite, a free coding agent model from JetBrains
- 21 Oct: Mistral's Voxtral 3: voice agents that cope with interruptions and switch languages
- 21 Oct: Claude ran most of a protein-design project by itself
- 21 Oct: Anthropic's IPO filing, as reported: revenue up fast, costs up faster
- 21 Oct: ChatGPT team tasks: recurring work your whole team can check
- 21 Oct: OpenAI's API can now keep your instructions cached for 12 hours
- 21 Oct: OpenAI buys Orchestra Bio, whose agents run drug-development operations
- 21 Oct: fx0: Vercel's free 'mini v0' for one-page prototypes
- 21 Oct: Treg's AEO Toolkit: see who AI recommends in your market
- 21 Oct: @steipete: a two-week agent loop rewrote his app's front end
- 21 Oct: @ctatedev: from one agent to a graph of agents that message each other
- 21 Oct: @Saboo_Shubham_: Google's free agent that watches your other agents
- 21 Oct: ego(lite): let Claude Code browse websites for you (a polished vendor promo)
- 21 Oct: Mike Futia's ScaleBot MCP: 12 ad concepts and images from one chat (his own product)
- 21 Oct: Mistral's Voxtral 3 launch demo: a voice agent that copes with real conversation
- 21 Oct: Check that AI search engines are actually allowed to read your site
- 21 Oct: @thsottiaux: Codex usage limits were wrong for about 40,000 accounts
- 21 Oct: @dr_cintas: Vyb, a design model that keeps your brand assets exactly as they are
- 20 Oct: MiniMax M3.1 Flash: a very cheap model for running agents, with one big caveat
- 20 Oct: The US orders DeepSeek out of federal agencies, including tools built on it
- 20 Oct: Anthropic, Google and OpenAI are all reportedly lining up new models
- 20 Oct: OpenAI says its newest models are much better at old business code
- 20 Oct: Meta explains how it trained its small open agent model, Muse Glimmer
- 20 Oct: A free library of about 50 marketing skills for your AI agent
- 20 Oct: Hy-Memory 2.0: one shared memory for all your AI coding tools
- 20 Oct: Boris Cherny on running dozens of agents at Anthropic (not a 'graph engineering' talk)
- 20 Oct: The 'Anthropic engineer builds a harness in 10 minutes' video is a 2025 Claude Code keynote
- 20 Oct: Greg Isenberg's 'one-person AI business' playbook: a decent idea, with big claims
- 20 Oct: Treat your review profiles and comparison pages as AI search assets
- 20 Oct: @dr_cintas: make images, video and music from inside Claude or ChatGPT with Sunra
- 20 Oct: @Sabrina_Ramonov: Instagram and Facebook DM automations, set up from ChatGPT
- 20 Oct: @mattpocockuk: his home-made AI wiki is now core to his business
- 20 Oct: @thsottiaux: OpenCode breach shows why API keys need limits
- 19 Oct: DeepSeek ships a faster, cheaper Flash model and opens its web search API
- 19 Oct: Anthropic puts 5% of its shares into a public benefit fund
- 19 Oct: Big brands are starting to block AI agents
- 19 Oct: Move step-by-step jobs out of CLAUDE.md and into skills
- 19 Oct: When your agent gets something wrong, save the reason as a skill
- 19 Oct: Inbox Zero Mail: an open-source alternative to Superhuman
- 19 Oct: shim-cli: stop your coding agent seeing your secrets
- 19 Oct: Letting AI agents work inside your ad accounts, safely
- 19 Oct: The viral "Andrew Ng graph engineering course" is really DeepLearning.AI's 2024 LangGraph course
- 19 Oct: Stripe's Head of Design on 'Zombie UI'
- 19 Oct: Check which of your pages Google's AI already shows
- 19 Oct: @kimmonismus: China drafts export rules for its biggest AI models
- 19 Oct: @kimmonismus and @Mr_Salio: three unreleased OpenAI models spotted
- 19 Oct: @Saboo_Shubham_: Google Research open-sources Harness2
- 19 Oct: @mattpocockuk: a one-prompt guardrail for AI-written code
- 19 Oct: @ctatedev: agent-browser now sends only what changed on a page
Hype or Fact? index
45 checks, verdict filter
Hype or Fact? check
45 pages, for example Grok Bot for LinkedIn: 'so well it feels illegal'.
All 45
- Grok Bot for LinkedIn: 'so well it feels illegal'
- Vercel CI: '2-6× faster CI', built for agents
- ScaleBot's review scraper: 'works with the top 10 review apps'
- Ploid: 'over 2 billion people indexed' for agents, 'for cents'
- Osaurus + LFM3: a 1.7B model 'read the whole case file' offline and cited its sources
- Supabase Compute: run agents next to your database
- ScaleBot Ad Spy: 'cancel Atria/Foreplay'
- Karpathy at AI Ascent: 'Opus 5.5' and '99% automated'
- Context.dev Parsers: paste a Google Maps, TikTok or G2 link, get clean data
- '18-year-old sold an agent to Anthropic for $3.4M'
- Supabase scoped access tokens: give your AI agent one project, not your whole account
- Google's Graph RAG codelab as a '$750,000 skill' course
- Gemini API CLI: Google's media models in one terminal command
- 'An AI pitch list in 18 minutes' for Main Street businesses
- A '$100k agent sold to Anthropic' post wrapped around a Stanford lecture
- '135 agents, 35 skills': the ready-made Claude Code teams clip
- HarnessRouter: '99.8% lower cost'
- Google's Agent102: 'a 2-hour course'
- CapCut in ChatGPT: 'coming soon'
- An AI motion graphics starter kit (Instagram reel)
- A Seedance 'Yapper' UGC ad, 'almost one shot'
- 'An engineer runs 65+ Claude agents in a loop'
- 'A SpaceX AI engineer runs 35+ bots overnight'
- Tech With Tim: a 40,000-post job tracker that 'costs nothing to run'
- Block's Dessa: a voice AI that listens while it talks
- A 'manipulated re-upload' of Jess Yan on long-running agents
- Supabase: give your app its own MCP server
- 1Password for AI agents: approve logins without handing over passwords
- 'Claude Code + Obsidian = $100,000 a month'
- ego(lite): 'the fastest browser for AI agents'
- ScaleBot MCP: 12 ad concepts and images from one chat
- Firecrawl People Data: free prospect research inside ChatGPT and Claude
- ElevenCreative Chimera: talking-character videos with expressive AI voices
- Claude Duet: Anthropic's proactive agent app
- fx0: Vercel's free 'mini v0' for one-page prototypes
- Treg's AEO Toolkit: see who AI recommends in your market
- Greg Isenberg's '$0' one-person AI business
- Boris Cherny: '85% of our engineers' and 'graph engineering'
- 'Anthropic engineer builds a harness in 10 minutes'
- '17-year-old runs 300 Grok Bots for under $0.02 a bot a week'
- The 'Andrew Ng graph engineering course'
- Google Chrome's three new tools for WebMCP
- 'Karpathy's second brain on Grok Bot'
- 'Set up a harness for your one-person company'
- 'Fully automated' AI SEO with an AI assistant
Worth testing index
32 items, grouped by day
Worth testing item
32 pages, for example Mike Futia: from Shopify reviews to Meta ad angles, inside ScaleBot.
All 32
- Mike Futia: from Shopify reviews to Meta ad angles, inside ScaleBot
- Claude for Legal: a regulatory horizon scan in a few clicks
- Mike Futia's ScaleBot Ad Spy: reading competitors' Meta ads
- Andrej Karpathy at Sequoia's AI Ascent: how to work with agents
- A Stanford lecture on 'more attempts plus a good checker'
- Greg Isenberg and 'Boring Marketer': how to sell AI to Main Street businesses
- Theo's tstack: the agent skills he uses to review and merge code
- Google's free Graph RAG agent codelab (reposted as a 'course')
- Jake Dabrowski's Claude Code workflow for product videos (dropshipping)
- A real Stanford lecture on LLMs and agents, wrapped in a fake '$100k agent' post
- The 'ready-made Claude Code teams' clip: what it really shows
- Lee Robinson's Stanford lecture: how always-on AI agents work
- Google's Agent102 lab: build a self-improving agent (not the '2-hour course' the post promises)
- HarnessRouter demo: one agent hands a data job to two specialist agents
- Mike Futia: a 'Yapper' talking-head UGC ad made with Seedance 2.5
- Anthropic's Lydia Hallie on cutting Claude Code costs (the post's '65 agents in a loop' claim is false)
- Cursor Cloud Agents webinar with Growth Engine X (don't trust the 'SpaceX AI engineer' framing)
- Tech With Tim: Claude Code builds a job-market tracker from 40,000+ Hacker News posts (sponsored by Databricks)
- Anthropic's Jess Yan on building long-running agents (a manipulated re-upload: watch Peter Yang's original instead)
- Greg Isenberg: getting your business ready for AI shopping agents (31-minute strategy talk)
- Nidhi Singh: let Claude Code use Codex's computer use to drive Mac apps (includes a sponsored segment)
- Claude Code + Obsidian: keep your agent's memory in a notes vault (likely a re-upload; credit @0xrux)
- ego(lite): let Claude Code browse websites for you (a polished vendor promo)
- Mike Futia's ScaleBot MCP: 12 ad concepts and images from one chat (his own product)
- Mistral's Voxtral 3 launch demo: a voice agent that copes with real conversation
- Boris Cherny on running dozens of agents at Anthropic (not a 'graph engineering' talk)
- The 'Anthropic engineer builds a harness in 10 minutes' video is a 2025 Claude Code keynote
- Greg Isenberg's 'one-person AI business' playbook: a decent idea, with big claims
- The viral "Andrew Ng graph engineering course" is really DeepLearning.AI's 2024 LangGraph course
- Stripe's Head of Design on 'Zombie UI'
- Google Cloud: building agent workflows as graphs with ADK 2.0
- An AI assistant SEO routine: research bot, writing skill and a human review
AI Radar
Interactive radar with tabs and method notes
About & methods
How we review, ratings, methods & sources, newsletter preview, corrections
Search
Search over tools, checks, worth testing and news
404 page
Not-found page
How we knowLogos: 238 tools show their maker’s own icon in ink tones; 40 without a usable icon get a lettered plate. Covers: the front page and news use the house-style covers; only 2 editor’s-pick tools get one.