[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"website-detail-dp-code":3,"facets-baseline-false":142,"facets-{\"includeAuthors\":\"false\"}":321,"similar-websites-dp-code":335},{"id":4,"name":5,"slug":6,"url":7,"dateUpdated":8,"fetchedAt":9,"coverId":10,"coverMobileId":11,"coverWidth":12,"coverHeight":13,"coverMobileWidth":14,"coverMobileHeight":15,"descriptionEn":16,"seoMeta":17,"industry":20,"styles":24,"credits":37,"pages":38,"typefaces":50,"technologiesByCategory":58,"pagespeedDesktop":88,"pagespeedMobile":100,"performanceDesktop":91,"performanceMobile":102,"colorBuckets":106,"colorPalette":123},"133670f9-6f06-4e62-b0f7-6ff1df286529","DP Code","dp-code","https:\u002F\u002Fwww.dpcode.cc","2026-05-23T19:23:53.000Z","2026-05-03T21:15:26.000Z","440174ba-3b47-4768-bf8e-011e6f6196bb","c9acd961-f45f-4a71-911f-0cfe0bcfceb3",1440,900,780,1688,"DP Code is a specialized developer tool designed to streamline the coding workflow by integrating various AI subscriptions into a single, cohesive environment. The website serves as a landing page for a productivity-focused software application that allows developers to leverage multiple AI models—such as Claude, ChatGPT, and others—simultaneously across different projects. Its primary purpose is to eliminate the friction of switching between multiple AI interfaces, offering a unified workspace for high-velocity software engineering.\n\nThe visual identity is defined by a sophisticated, ultra-minimalist aesthetic that leans heavily into a light-mode, clean design language. Utilizing ample whitespace, precise typography, and subtle UI mockups, the site communicates efficiency and technical clarity. The target audience consists of professional software engineers and power users who demand high-performance tools. The design avoids unnecessary ornamentation, opting instead for a functional, high-end software-as-a-service (SaaS) look that emphasizes ease of use and professional reliability.",{"title":18,"meta_description":19},"DP Code | Code with Your AI Subscriptions Seamlessly","Maximize productivity with DP Code. The ultimate workspace to use all your AI subscriptions like Claude and ChatGPT in one unified coding environment.",{"name":21,"label":22,"slug":23},"SaaS & Software","Software","software",[25,29,33],{"name":26,"label":27,"slug":28},"Corporate \u002F Professional","Corporate","corporate",{"name":30,"label":31,"slug":32},"Clean \u002F Minimalist","Minimalist","minimalist",{"name":34,"label":35,"slug":36},"Monochrome \u002F Grayscale","Monochrome","monochrome",[],[39],{"name":40,"title":41,"description":42,"url":43,"coverId":10,"coverMobileId":11,"alt":44,"altMobile":45,"pageType":46},"Homepage","DP Code - The best way to code with your AI subscriptions","DP Code - the best way to code with your AI subscriptions.","https:\u002F\u002Fwww.dpcode.cc\u002F","The desktop homepage for DP Code features a clean, minimalist white layout with dark text. A hero section includes the headline 'Built to make you extraordinarily productive,' primary buttons for 'Download for Linux' and 'Star on GitHub,' and two side-by-side screenshots showing the software interface in light and dark modes. Below, a 'Providers' section lists various AI subscriptions like Claude and Codex using a grid of text and icons.","The mobile version presents a vertically stacked layout optimized for small screens. The hero headline and download buttons are centered, and the side-by-side interface screenshots from the desktop version are replaced by a single, large interface preview. The 'Providers' section and subsequent feature details follow a single-column scrolling format.",{"name":47,"label":48,"slug":49},"Home \u002F Landing Page","Home \u002F Landing","home-landing",[51,55],{"name":52,"label":52,"slug":53,"foundryName":54},"Geist","geist",null,{"name":56,"label":56,"slug":57,"foundryName":54},"Geist Mono","geist-mono",{"Security":59,"CDN":63,"JavaScript frameworks":67,"Web frameworks":74,"Programming languages":76,"PaaS":80,"Miscellaneous":84},[60],{"name":61,"label":61,"slug":62},"HSTS","hsts",[64],{"name":65,"label":65,"slug":66},"jsDelivr","jsdelivr",[68,71],{"name":69,"label":69,"slug":70},"Next.js","nextjs",{"name":72,"label":72,"slug":73},"React","react",[75],{"name":69,"label":69,"slug":70},[77],{"name":78,"label":78,"slug":79},"Node.js","nodejs",[81],{"name":82,"label":82,"slug":83},"Vercel","vercel",[85],{"name":86,"label":86,"slug":87},"Webpack","webpack",[89,92,95,98],{"category":90,"score":91},"performance",98,{"category":93,"score":94},"accessibility",79,{"category":96,"score":97},"best-practices",100,{"category":99,"score":97},"seo",[101,103,104,105],{"category":90,"score":102},80,{"category":93,"score":94},{"category":96,"score":97},{"category":99,"score":97},[107,111,115,119],{"value":108,"label":109,"hex":110},"white","White","#ffffff",{"value":112,"label":113,"hex":114},"gray","Gray","#737373",{"value":116,"label":117,"hex":118},"black","Black","#000000",{"value":120,"label":121,"hex":122},"green","Green","#00c950",[124,127,130,133,136,139],{"hex":125,"percent":126},"#F3F3F1",0.9365,{"hex":128,"percent":129},"#CFCFD3",0.0183,{"hex":131,"percent":132},"#9A9EB2",0.0161,{"hex":134,"percent":135},"#1D1C1B",0.0157,{"hex":137,"percent":138},"#696968",0.0134,{"hex":140,"percent":141},"#38C360",0.0001,{"facetDistribution":143,"facetLabels":229},{"industry":144,"styles":164,"technologies":179,"font_families":204,"page_type_name":218},{"Agency & Studio":145,"Art & Culture":146,"SaaS & Software":147,"Fashion & Apparel":148,"Architecture & Real Estate":149,"Entertainment & Media":150,"Food & Beverage":151,"Photography & Video":152,"Finance & Fintech":153,"Healthcare & Wellness":154,"Technology & Hardware":155,"Industrial & Manufacturing":156,"Beauty & Cosmetics":157,"Travel & Hospitality":158,"Non-Profit & Charity":159,"Education & Courses":160,"Crypto & Web3":161,"Sports & Fitness":162,"Automotive & Transportation":163},716,296,210,167,163,138,125,120,76,62,54,49,47,45,41,32,24,21,11,{"Clean \u002F Minimalist":165,"Typographic \u002F Big Type":166,"High-End \u002F Luxury":167,"Dark Mode":168,"Vibrant \u002F Colorful":169,"Monochrome \u002F Grayscale":170,"Fun \u002F Playful":171,"Brutalist \u002F Neo-Brutalist":172,"Corporate \u002F Professional":173,"Illustrative \u002F Hand-drawn":174,"Futuristic \u002F Sci-Fi":175,"3D \u002F Spatial":176,"Retro \u002F Vintage \u002F Y2K":177,"Glassmorphism":178,"Bento Grid":178},2056,1713,1012,570,393,317,257,236,205,87,69,55,33,15,{"jQuery":180,"PHP":181,"WordPress":182,"Node.js":183,"React":184,"Vercel":185,"Vue.js":186,"Shopify":187,"Next.js":188,"Klaviyo":189,"Nuxt.js":190,"Typekit":191,"Amazon Web Services":192,"reCAPTCHA":193,"YouTube":194,"MailChimp":195,"Modernizr":196,"Bootstrap":197,"Slick":198,"Google Font API":94,"Craft CMS":199,"Framer Sites":200,"Contact Form 7":201,"Stimulus":202,"Google Maps":203},875,629,504,320,311,262,207,196,161,149,147,141,129,117,95,92,91,85,82,66,65,64,59,58,{"Inter":205,"Times New Roman":206,"Roboto":207,"Open Sans":208,"Neue Haas Grotesk":208,"Suisse Intl":209,"Arial":175,"Poppins":199,"Helvetica Neue":201,"Montserrat":210,"Diatype":176,"Neue Montreal":155,"Lato":156,"Graphik":211,"Source Sans":212,"Monument Grotesk":213,"Noto Sans":214,"Helvetica":215,"Söhne":216,"DM Sans":216,"Founders Grotesk":216,"GT America":217,"Google Sans":217,"Editorial":177,"IBM Plex Mono":160},315,168,127,90,88,60,46,44,40,39,36,35,34,{"Home \u002F Landing Page":219,"About Us \u002F Team":220,"Projects \u002F Portfolio":221,"Contact Us":222,"Blog \u002F Article Layout":223,"Services \u002F Features":224,"Shop \u002F Catalog":225,"Careers \u002F Jobs":226,"Pricing Page":154,"Sign Up \u002F Login":227,"404 Error Page":228},2401,774,527,418,378,341,299,89,27,9,{"industry":230,"styles":249,"page_type_name":262,"technologies":273,"font_families":295},{"Agency & Studio":231,"Art & Culture":232,"SaaS & Software":22,"Fashion & Apparel":233,"Architecture & Real Estate":234,"Entertainment & Media":235,"Food & Beverage":236,"Photography & Video":237,"Finance & Fintech":238,"Healthcare & Wellness":239,"Technology & Hardware":240,"Industrial & Manufacturing":241,"Beauty & Cosmetics":242,"Travel & Hospitality":243,"Non-Profit & Charity":244,"Education & Courses":245,"Crypto & Web3":246,"Sports & Fitness":247,"Automotive & Transportation":248},"Agencies","Art and Culture","Fashion","Real Estate","Entertainment","Food and Drink","Photography","Fintech","Wellness","Technology","Industrial","Beauty","Travel","Non-Profit","Education","Web3","Sports","Automotive",{"Clean \u002F Minimalist":31,"Typographic \u002F Big Type":250,"High-End \u002F Luxury":251,"Dark Mode":252,"Vibrant \u002F Colorful":253,"Monochrome \u002F Grayscale":35,"Fun \u002F Playful":254,"Brutalist \u002F Neo-Brutalist":255,"Corporate \u002F Professional":27,"Illustrative \u002F Hand-drawn":256,"Futuristic \u002F Sci-Fi":257,"3D \u002F Spatial":258,"Retro \u002F Vintage \u002F Y2K":259,"Glassmorphism":260,"Bento Grid":261},"Typographic","Luxury","Dark Mode","Vibrant","Playful","Brutalist","Illustrative","Futuristic","3D and Spatial","Retro and Y2K","Glassmorphism","Bento Grid",{"Home \u002F Landing Page":48,"About Us \u002F Team":263,"Projects \u002F Portfolio":264,"Contact Us":265,"Blog \u002F Article Layout":266,"Services \u002F Features":267,"Shop \u002F Catalog":268,"Careers \u002F Jobs":269,"Pricing Page":270,"Sign Up \u002F Login":271,"404 Error Page":272},"About Us","Portfolio","Contact","Blog \u002F Article","Services","Shop \u002F Catalog","Careers","Pricing","Sign Up \u002F Login","404 Error",{"jQuery":274,"PHP":275,"WordPress":276,"Node.js":78,"React":72,"Vercel":82,"Vue.js":277,"Shopify":278,"Next.js":69,"Klaviyo":279,"Nuxt.js":280,"Typekit":281,"Amazon Web Services":282,"reCAPTCHA":283,"YouTube":284,"MailChimp":285,"Modernizr":286,"Bootstrap":287,"Slick":288,"Google Font API":289,"Craft CMS":290,"Framer Sites":291,"Contact Form 7":292,"Stimulus":293,"Google Maps":294},"jQuery","PHP","WordPress","Vue.js","Shopify","Klaviyo","Nuxt.js","Typekit","Amazon Web Services","reCAPTCHA","YouTube","MailChimp","Modernizr","Bootstrap","Slick","Google Font API","Craft CMS","Framer Sites","Contact Form 7","Stimulus","Google Maps",{"Inter":296,"Times New Roman":297,"Roboto":298,"Open Sans":299,"Neue Haas Grotesk":300,"Suisse Intl":301,"Arial":302,"Poppins":303,"Helvetica Neue":304,"Montserrat":305,"Diatype":306,"Neue Montreal":307,"Lato":308,"Graphik":309,"Source Sans":310,"Monument Grotesk":311,"Noto Sans":312,"Helvetica":313,"Söhne":314,"DM Sans":315,"Founders Grotesk":316,"GT America":317,"Google Sans":318,"Editorial":319,"IBM Plex Mono":320},"Inter","Times New Roman","Roboto","Open Sans","Neue Haas Grotesk","Suisse Intl","Arial","Poppins","Helvetica Neue","Montserrat","Diatype","Neue Montreal","Lato","Graphik","Source Sans","Monument Grotesk","Noto Sans","Helvetica","Söhne","DM Sans","Founders Grotesk","GT America","Google Sans","Editorial","IBM Plex Mono",{"facetDistribution":322,"facetLabels":329},{"industry":323,"styles":324,"technologies":325,"font_families":326,"page_type_name":327},{"Agency & Studio":145,"Art & Culture":146,"SaaS & Software":147,"Fashion & Apparel":148,"Architecture & Real Estate":149,"Entertainment & Media":150,"Food & Beverage":151,"Photography & Video":152,"Finance & Fintech":153,"Healthcare & Wellness":154,"Technology & Hardware":155,"Industrial & Manufacturing":156,"Beauty & Cosmetics":157,"Travel & Hospitality":158,"Non-Profit & Charity":159,"Education & Courses":160,"Crypto & Web3":161,"Sports & Fitness":162,"Automotive & Transportation":163},{"Clean \u002F Minimalist":165,"Typographic \u002F Big Type":166,"High-End \u002F Luxury":167,"Dark Mode":168,"Vibrant \u002F Colorful":169,"Monochrome \u002F Grayscale":170,"Fun \u002F Playful":171,"Brutalist \u002F Neo-Brutalist":172,"Corporate \u002F Professional":173,"Illustrative \u002F Hand-drawn":174,"Futuristic \u002F Sci-Fi":175,"3D \u002F Spatial":176,"Retro \u002F Vintage \u002F Y2K":177,"Glassmorphism":178,"Bento Grid":178},{"jQuery":180,"PHP":181,"WordPress":182,"Node.js":183,"React":184,"Vercel":185,"Vue.js":186,"Shopify":187,"Next.js":188,"Klaviyo":189,"Nuxt.js":190,"Typekit":191,"Amazon Web Services":192,"reCAPTCHA":193,"YouTube":194,"MailChimp":195,"Modernizr":196,"Bootstrap":197,"Slick":198,"Google Font API":94,"Craft CMS":199,"Framer Sites":200,"Contact Form 7":201,"Stimulus":202,"Google Maps":203},{"Inter":205,"Times New Roman":206,"Roboto":207,"Open Sans":208,"Neue Haas Grotesk":208,"Suisse Intl":209,"Arial":175,"Poppins":199,"Helvetica Neue":201,"Montserrat":210,"Diatype":176,"Neue Montreal":155,"Lato":156,"Graphik":211,"Source Sans":212,"Monument Grotesk":213,"Noto Sans":214,"Helvetica":215,"Söhne":216,"DM Sans":216,"Founders Grotesk":216,"GT America":217,"Google Sans":217,"Editorial":177,"IBM Plex Mono":160},{"Home \u002F Landing Page":328},2397,{"industry":330,"styles":331,"page_type_name":332,"technologies":333,"font_families":334},{"Agency & Studio":231,"Art & Culture":232,"SaaS & Software":22,"Fashion & Apparel":233,"Architecture & Real Estate":234,"Entertainment & Media":235,"Food & Beverage":236,"Photography & Video":237,"Finance & Fintech":238,"Healthcare & Wellness":239,"Technology & Hardware":240,"Industrial & Manufacturing":241,"Beauty & Cosmetics":242,"Travel & Hospitality":243,"Non-Profit & Charity":244,"Education & Courses":245,"Crypto & Web3":246,"Sports & Fitness":247,"Automotive & Transportation":248},{"Clean \u002F Minimalist":31,"Typographic \u002F Big Type":250,"High-End \u002F Luxury":251,"Dark Mode":252,"Vibrant \u002F Colorful":253,"Monochrome \u002F Grayscale":35,"Fun \u002F Playful":254,"Brutalist \u002F Neo-Brutalist":255,"Corporate \u002F Professional":27,"Illustrative \u002F Hand-drawn":256,"Futuristic \u002F Sci-Fi":257,"3D \u002F Spatial":258,"Retro \u002F Vintage \u002F Y2K":259,"Glassmorphism":260,"Bento Grid":261},{"Home \u002F Landing Page":48},{"jQuery":274,"PHP":275,"WordPress":276,"Node.js":78,"React":72,"Vercel":82,"Vue.js":277,"Shopify":278,"Next.js":69,"Klaviyo":279,"Nuxt.js":280,"Typekit":281,"Amazon Web Services":282,"reCAPTCHA":283,"YouTube":284,"MailChimp":285,"Modernizr":286,"Bootstrap":287,"Slick":288,"Google Font API":289,"Craft CMS":290,"Framer Sites":291,"Contact Form 7":292,"Stimulus":293,"Google Maps":294},{"Inter":296,"Times New Roman":297,"Roboto":298,"Open Sans":299,"Neue Haas Grotesk":300,"Suisse Intl":301,"Arial":302,"Poppins":303,"Helvetica Neue":304,"Montserrat":305,"Diatype":306,"Neue Montreal":307,"Lato":308,"Graphik":309,"Source Sans":310,"Monument Grotesk":311,"Noto Sans":312,"Helvetica":313,"Söhne":314,"DM Sans":315,"Founders Grotesk":316,"GT America":317,"Google Sans":318,"Editorial":319,"IBM Plex Mono":320},{"hits":336},[337,472,602,734],{"id":338,"website_id":339,"page_id":338,"name":340,"slug":341,"url":342,"website_name":340,"website_slug":341,"website_url":343,"result_url":342,"fetched_at":344,"score":345,"score_boost":346,"ai_score":347,"freshness_score":348,"scored_at":349,"page_name":40,"page_url":342,"page_title":350,"page_description":351,"page_content":352,"page_sort":346,"is_home":353,"is_home_rank":354,"page_type_id":355,"page_type_name":47,"cover":356,"cover_mobile":359,"cover_sequence":361,"translations":449,"industry":21,"styles":453,"credits":455,"font_families":456,"technologies":457,"pagespeed":458,"pagespeed_mobile":464,"buckets":470,"search_payload":471},"f64a99d6-8132-4221-9005-413dc69c40fb","54ab08c3-d78d-4c08-ad82-b351209b2e3b","Superset","superset","https:\u002F\u002Fsuperset.sh\u002F","https:\u002F\u002Fsuperset.sh","2026-05-03T21:16:35.000Z",81,0,83,74,"2026-05-23T19:17:32.000Z","Superset - Run 10+ parallel coding agents on your machine","Run 10+ parallel coding agents on your machine. Spin up new coding tasks while waiting for your current agent to finish. Quickly switch between tasks as they need your attention.","The Code Editor for AI Agents.The Cod Orchestrate 100+ coding agents in parallel. Works for any agents. Built for the AI era. Join WaitlistView on GitHub superset New Workspace superset(5) ⠼ new workspace creating... ⠼ use any agents +46\\-1 use-any-agents#733 create parallel branches +193 create-parallel-branches#815 see changes +394\\-23 see-changes#884 open in any IDE +33 open-in-any-ide#816 forward ports +127\\-8 forward-ports#902 ⌥Ports 4 use any agents 3002 see changes 300030015678 Claudeclaude Codexcodex Geminigemini Cursorcursor ⬛Terminal □ \\ ▐▛███▜▌ \\ \\ ▝▜█████▛▘ \\ \\ ▘▘ ▝▝ \\ Claude Code v2.0.74 Opus 4.5 · Claude Max ~\u002F.superset\u002Fworktrees\u002Fsuperset\u002Fcloud-ws ❯ \u002Fmcp MCP Servers 1 connected ❯1.superset-mcp✓ connected config: .mcp.json ❯Type a task for Claude... ↑ ❯ superset new ⠼Setting up new parallel environment... → create worktree → install deps → ready shell Review Changes #827 Commit message... ↑Push26 bun.lock +38\\-25 packages\u002Fdb\u002Fsrc\u002Fschema cloud-workspace.ts +119 enums.ts +21 apps\u002Fdesktop\u002Fsrc\u002Frenderer CloudTerminal.tsx +169 useCloudWorkspaces.ts +84 LegacyTerminalPane.tsx \\-42 WorkspaceSidebar.tsx +14 apps\u002Fapi\u002Fsrc\u002Ftrpc\u002Frouters ssh-manager.ts +277 index.ts +7 Review PR #827 Open cloud-workspace.tsenums.ts+4 more @@ -1,4 +1,6 @@ 1import { db } from \"..\u002Fdb\" +import { CloudWorkspace } from \".\u002Ftypes\" +import { createSSHConnection } from \".\u002Fssh\" 2 \\-export const getWorkspaces = () =\\> { +export const getWorkspaces = async () =\\> { 4 return db.query.workspaces ApproveComment External IDE src index.ts utils.ts types.ts import { Agent } from \"ai\" import { tools } from \".\u002Futils\" │ const agent = new Agent({ model: \"claude-4\", tools: \\[tools.read, tools.write\\] }) Use Any AgentsCreate Parallel BranchesSee ChangesOpen in Any IDE Trusted by builders from Microsoft OpenAI Netflix Wordware Salesforce Wix Datadog Intercom ByteDance Toss Google Vercel Cloudflare Amazon Microsoft OpenAI Netflix Wordware Salesforce Wix Datadog Intercom ByteDance Toss Google Vercel Cloudflare Amazon Parallel Execution Run dozens of agents at once Launch multiple AI coding agents across different tasks. Work on features, fix bugs, and refactor code — all in parallel. Superset In Progress 3 Analyze Tab vs Agent Usag... Generating PyTorch MNIST Experiments Generating Fix PR Comments Fetching I... Generating Ready for Review 3 Enterprise Order Mana...now +93\\-18· Perfect! I've implem... Set up Cursor Rules fo...30m +37· Set up Cursor Rules f... Bioinformatics Tools45m +135\\-21· Bioinformatics Tools \\> ⌘↵ Universal Compatibility Works with any CLI agent Superset is agent-agnostic. Use Claude Code, OpenCode, Cursor, or any CLI-based coding tool. Switch between agents seamlessly. superset main New Terminal... Claude Claude OpenCode OpenCode Codex Codex Gemini Gemini Cursor Agent Cursor Agent Terminals (3) Isolation Changes are isolated Each agent runs in its own isolated Git worktree. No merge conflicts, no stepping on each other's changes. Review and merge work when you're ready. components\u002FHeroSection\u002FHeroSection.tsx Side by SideInline Messages Commits3 Against Main Unstaged HeroSection.tsx+12\\-3 GridBackground.ts+45 constants.ts+8\\-2 ProductDemo.tsx+23\\-15 1 export function HeroSection() { 2 return ( 3\\- \u003Cdiv className=\"hero-old\"> 4+ \u003Csection className=\"relative py-24\"> 5+ \u003CGridBackground \u002F> 6 \u003Cdiv className=\"max-w-7xl mx-auto\"> 7\\- \u003Ch1>Welcome\u003C\u002Fh1> 8+ \u003Cmotion.h1 9+ initial={{ opacity: 0 }} 10+ animate={{ opacity: 1 }} 11+ > 12+ Superset 13+ \u003C\u002Fmotion.h1> 14 \u003C\u002Fdiv> 15\\- \u003C\u002Fdiv> 16+ \u003C\u002Fsection> 17 ); 18 } Open Anywhere Open in any IDE Jump into your favorite editor with one click. VS Code, Cursor, Xcode, JetBrains IDEs, or any terminal — open worktrees exactly where you need them. superset src Search files... CursorOpen in components HeroSection.tsx constants.ts index.ts hooks utils Finder Finder Cursor Cursor ⌘O VS Code VS Code Xcode Xcode Sublime Text Sublime Text Terminal Terminal JetBrains JetBrains What builders say about Superset [ Abhi Aiyer Abhi Aiyer Co-founder & CTO at Mastra Just realized that I have done all my work in @superset\\_sh since Dec 26. ](https:\u002F\u002Fx.com\u002Fabhiaiyer\u002Fstatus\u002F2013782002332283180) [ Zach Dive Zach Dive Co-founder & CEO at Adam if you're not using @superset\\_sh, you're getting left behind in 2026 ](https:\u002F\u002Fx.com\u002Fzachdive\u002Fstatus\u002F2014038312508424597) [ Chase McDougall Chase McDougall Founding Engineer at Decoda Health @superset\\_sh is a sick product - love OS since I don't have to wait for someone else to fix bugs ](https:\u002F\u002Fx.com\u002FChaseMcDou\u002Fstatus\u002F2013458004977373643) [ Vlad Arbatov Vlad Arbatov Founding Engineer at Loyal superset became my default tools now so keep the great work folks ](https:\u002F\u002Fx.com\u002Fvladzima\u002Fstatus\u002F2032306550073610246) [ Chris Laupama Chris Laupama TS Lead at Webflow Oh snap @superset\\_sh is soooooo damn good! Now THIS is the experience I've been wanting for CLI agents! ](https:\u002F\u002Fx.com\u002Fchrislaupama\u002Fstatus\u002F2011148329443607037) [ Eric Clemmons Eric Clemmons Principal Engineer at Cloudflare Was using Warp, but now @superset\\_sh has become my primary terminal ](https:\u002F\u002Fx.com\u002Fericclemmons\u002Fstatus\u002F2013413118467056004) [ Leo Leo Co-founder & CTO at Outlit hasn't been a day i haven't used superset since onboarding complete paradigm shift on how i use ai to code ](https:\u002F\u002Fx.com\u002FLeosReal\u002Fstatus\u002F2027306293858586745) [ Elias Ståvik Elias Ståvik Founder at Cleanroom just started using remote desktop instead of ssh just to be able to use @superset\\_sh on my mac mini from my macbook pro generational product in the making, mark my words ](https:\u002F\u002Fx.com\u002Feliasstravik\u002Fstatus\u002F2020580091449708978) [ Gregor Zunic Gregor Zunic Co-founder & CTO at Browser Use Damn @superset\\_sh is really cool. You should try it. worktrees are a breeze cmd + t auto opens Claude Code you can view git changes within itself closing a laptop doesn't kill the sessions ](https:\u002F\u002Fx.com\u002Fgregpr07\u002Fstatus\u002F2013038355630432742) [ Felipe Coury Felipe Coury Codex at OpenAI If you prefer a more GUI-oriented approach to multiple agents in parallel, it seems like @superset\\_sh is doing a tremendous job. ](https:\u002F\u002Fx.com\u002Ffcoury\u002Fstatus\u002F2010477904472281220) [ Iven Iven Engineer at Paraflow Tested various GUI tools for git worktree and agents - Conductor, Vibe Kanban, Agentastic, Crystal, FleetCode, Emdash, Sculptor - but Superset suits my taste the best TranslatedShow original](https:\u002F\u002Fx.com\u002Fivenvd\u002Fstatus\u002F2011738469610242559) Frequently asked questions I already use an IDE like Cursor, is this for me? Which AI coding agents are supported? How does the parallel agent system work? Is Superset free to use? Can I use my own API keys? Try Superset now. Join Waitlist",true,1,"1",{"id":357,"height":13,"width":12,"blurhash":358},"daa6f2d4-c200-4783-ba52-b78f4d81456a","radial-gradient(at 0 0,#201d1d,#00000000 50%),radial-gradient(at 33% 0,#232324,#00000000 50%),radial-gradient(at 67% 0,#272929,#00000000 50%),radial-gradient(at 100% 0,#222322,#00000000 50%),radial-gradient(at 0 50%,#1b1d1e,#00000000 50%),radial-gradient(at 33% 50%,#323232,#00000000 50%),radial-gradient(at 67% 50%,#353434,#00000000 50%),radial-gradient(at 100% 50%,#242424,#00000000 50%),radial-gradient(at 0 100%,#181817,#00000000 50%),radial-gradient(at 33% 100%,#2c2c2c,#00000000 50%),radial-gradient(at 67% 100%,#292929,#00000000 50%),radial-gradient(at 100% 100%,#171716,#00000000 50%)",{"id":360,"height":15,"width":14,"blurhash":54},"74dbe250-b019-4e6b-9ace-04b17b82af38",[362,365,368,371,374,377,380,383,386,389,392,395,398,401,404,407,410,413,416,419,422,425,428,431,434,437,440,443,446],{"directus_files_id":363},{"id":364},"9c9221d0-92fa-43f9-8ad4-d013ec7b3ae6",{"directus_files_id":366},{"id":367},"8deb10a7-05e8-45c2-88db-0b5aa29223f9",{"directus_files_id":369},{"id":370},"debb819c-e4f3-401e-ba27-6c9c6fad197f",{"directus_files_id":372},{"id":373},"b4c7b110-4818-4ebc-bab8-22e4d50ab55d",{"directus_files_id":375},{"id":376},"dc38a8ad-9ddd-4795-bb55-3c656aad5e2f",{"directus_files_id":378},{"id":379},"cb6ab95f-0636-4c12-afd5-25ba51d76e30",{"directus_files_id":381},{"id":382},"07f93bcb-4966-40ff-944b-72769df84701",{"directus_files_id":384},{"id":385},"41d9dfb3-c621-452a-ba72-1663fca67b84",{"directus_files_id":387},{"id":388},"d7e56070-3699-48fb-8909-d0188afd5d43",{"directus_files_id":390},{"id":391},"e381b610-50fd-4cf3-bf8f-8c93e1f09ada",{"directus_files_id":393},{"id":394},"aecb883b-1bab-434c-bb4b-e51b20f9a24e",{"directus_files_id":396},{"id":397},"b4aed0e0-1512-419e-be5e-219c0f4dcf3a",{"directus_files_id":399},{"id":400},"62e04cf4-3d2e-436b-8214-fbc3aff6fa75",{"directus_files_id":402},{"id":403},"485d89fb-314d-4e7f-97d0-a3c28b4dd976",{"directus_files_id":405},{"id":406},"70597c13-b029-467f-80c0-ad2d894ff730",{"directus_files_id":408},{"id":409},"276d402d-0243-4b95-a6a6-c3c8cd5b75fd",{"directus_files_id":411},{"id":412},"98d9f746-7746-4046-b570-b1b165c089ce",{"directus_files_id":414},{"id":415},"573125dd-7f23-4736-90bd-feb84f82b43d",{"directus_files_id":417},{"id":418},"9ee8166a-14f1-4308-8a58-474e569ed85b",{"directus_files_id":420},{"id":421},"bec5b238-85d9-44c5-a76b-ebd0e350c2a8",{"directus_files_id":423},{"id":424},"73a1132b-97f1-4ad5-b1ad-98dfb4a4a638",{"directus_files_id":426},{"id":427},"1fdb6c8e-b645-456e-a516-a85d2a5592f0",{"directus_files_id":429},{"id":430},"9be97477-98e6-4092-aae8-5c6286b24fd5",{"directus_files_id":432},{"id":433},"9b2c2fae-8cc5-4002-bed5-e5b137fdd039",{"directus_files_id":435},{"id":436},"2bf6569a-1123-4cb2-942a-72db769cfd55",{"directus_files_id":438},{"id":439},"b334a0d1-8dee-4d47-ab69-1a653b1c6a91",{"directus_files_id":441},{"id":442},"9cecb649-7b04-4f93-8705-bff065ff2e74",{"directus_files_id":444},{"id":445},"7284b979-0062-472b-8077-84c23c0704a1",{"directus_files_id":447},{"id":448},"81c7099c-70c1-4fc6-bb25-cf60dd0351ea",[450],{"languages_code":451,"description":452},"en","Superset is a high-performance developer tool designed to orchestrate and run multiple AI coding agents in parallel. Positioned as \"The Code Editor for AI Agents,\" the platform aims to maximize developer productivity by allowing users to spin up parallel branches of coding tasks, effectively eliminating the downtime typically spent waiting for single-agent execution. It is built for the next generation of software engineers who integrate autonomous agents into their local development workflows.\n\nThe visual identity is unapologetically technical and modern, utilizing a sophisticated dark mode aesthetic that resonates with the developer community. The design employs heavy, impactful typography and a structured grid layout to showcase social proof from industry giants like Microsoft, OpenAI, and Netflix. By combining a minimalist color palette with high-contrast elements and a clean, functional interface, Superset establishes a sense of power, efficiency, and cutting-edge technological authority.",[30,252,454],"Typographic \u002F Big Type",[],[320,296],[69,78,72,82],[459,461,462,463],{"score":460,"category":90},57,{"score":97,"category":93},{"score":97,"category":96},{"score":97,"category":99},[465,467,468,469],{"score":466,"category":90},30,{"score":97,"category":93},{"score":97,"category":96},{"score":97,"category":99},[116,108,112],"Website: Superset. Page: Homepage. Page type: Home \u002F Landing Page. Page title: Superset - Run 10+ parallel coding agents on your machine. Page description: Run 10+ parallel coding agents on your machine. Spin up new coding tasks while waiting for your current agent to finish. Quickly switch between tasks as they need your attention.. Page content: The Code Editor for AI Agents.The Cod Orchestrate 100+ coding agents in parallel. Works for any agents. Built for the AI era. Join WaitlistView on GitHub superset New Workspace superset(5) ⠼ new workspace creating... ⠼ use any agents +46\\-1 use-any-agents#733 create parallel branches +193 create-parallel-branches#815 see changes +394\\-23 see-changes#884 open in any IDE +33 open-in-any-ide#816 forward ports +127\\-8 forward-ports#902 ⌥Ports 4 use any agents 3002 see changes 300030015678 Claudeclaude Codexcodex Geminigemini Cursorcursor ⬛Terminal □ \\ ▐▛███▜▌ \\ \\ ▝▜█████▛▘ \\ \\ ▘▘ ▝▝ \\ Claude Code v2.0.74 Opus 4.5 · Claude Max ~\u002F.superset\u002Fworktrees\u002Fsuperset\u002Fcloud-ws ❯ \u002Fmcp MCP Servers 1 connected ❯1.superset-mcp✓ connected config: .mcp.json ❯Type a task for Claude... ↑ ❯ superset new ⠼Setting up new parallel environment... → create worktree → install deps → ready shell Review Changes #827 Commit message... ↑Push26 bun.lock +38\\-25 packages\u002Fdb\u002Fsrc\u002Fschema cloud-workspace.ts +119 enums.ts +21 apps\u002Fdesktop\u002Fsrc\u002Frenderer CloudTerminal.tsx +169 useCloudWorkspaces.ts +84 LegacyTerminalPane.tsx \\-42 WorkspaceSidebar.tsx +14 apps\u002Fapi\u002Fsrc\u002Ftrpc\u002Frouters ssh-manager.ts +277 index.ts +7 Review PR #827 Open cloud-workspace.tsenums.ts+4 more @@ -1,4 +1,6 @@ 1import { db } from \"..\u002Fdb\" +import { CloudWorkspace } from \".\u002Ftypes\" +import { createSSHConnection } from \".\u002Fssh\" 2 \\-export const getWorkspaces = () =\\> { +export const getWorkspaces = async () =\\> { 4 return db.query.workspaces ApproveComment External IDE src index.ts utils.ts types.ts import { Agent } from \"ai\" import { tools } from \".\u002Futils\" │ const agent = new Agent({ model: \"claude-4\", tools: \\[tools.read, tools.write\\] }) Use Any AgentsCreate Parallel BranchesSee ChangesOpen in Any IDE Trusted by builders from Microsoft OpenAI Netflix Wordware Salesforce Wix Datadog Intercom ByteDance Toss Google Vercel Cloudflare Amazon Microsoft OpenAI Netflix Wordware Salesforce Wix Datadog Intercom ByteDance Toss Google Vercel Cloudflare Amazon Parallel Execution Run dozens of agents at once Launch multiple AI coding agents across different tasks. Work on features, fix bugs, and refactor code — all in parallel. Superset In Progress 3 Analyze Tab vs Agent Usag... Generating PyTorch MNIST Experiments Generating Fix PR Comments Fetching I... Generating Ready for Review 3 Enterprise Order Mana...now +93\\-18· Perfect! I've implem... Set up Cursor Rules fo...30m +37· Set up Cursor Rules f... Bioinformatics Tools45m +135\\-21· Bioinformatics Tools \\> ⌘↵ Universal Compatibility Works with any CLI agent Superset is agent-agnostic. Use Claude Code, OpenCode, Cursor, or any CLI-based coding tool. Switch between agents seamlessly. superset main New Terminal... Claude Claude OpenCode OpenCode Codex Codex Gemini Gemini Cursor Agent Cursor Agent Terminals (3) Isolation Changes are isolated Each agent runs in its own isolated Git worktree. No merge conflicts, no stepping on each other's changes. Review and merge work when you're ready. components\u002FHeroSection\u002FHeroSection.tsx Side by SideInline Messages Commits3 Against Main Unstaged HeroSection.tsx+12\\-3 GridBackground.ts+45 constants.ts+8\\-2 ProductDemo.tsx+23\\-15 1 export function HeroSection() { 2 return ( 3\\- \u003Cdiv className=\"hero-old\"> 4+ \u003Csection className=\"relative py-24\"> 5+ \u003CGridBackground \u002F> 6 \u003Cdiv className=\"max-w-7xl mx-auto\"> 7\\- \u003Ch1>Welcome\u003C\u002Fh1> 8+ \u003Cmotion.h1 9+ initial={{ opacity: 0 }} 10+ animate={{ opacity: 1 }} 11+ > 12+ Superset 13+ \u003C\u002Fmotion.h1> 14 \u003C\u002Fdiv> 15\\- \u003C\u002Fdiv> 16+ \u003C\u002Fsection> 17 ); 18 } Open Anywhere Open in any IDE Jump into your favorite editor with one click. VS Code, Cursor, Xcode, JetBrains IDEs, or any terminal — open worktrees exactly where you need them. superset src Search files... CursorOpen in components HeroSection.tsx constants.ts index.ts hooks utils Finder Finder Cursor Cursor ⌘O VS Code VS Code Xcode Xcode Sublime Text Sublime Text Terminal Terminal JetBrains JetBrains What builders say about Superset [ Abhi Aiyer Abhi Aiyer Co-founder & CTO at Mastra Just realized that I have done all my work in @superset\\_sh since Dec 26. ](https:\u002F\u002Fx.com\u002Fabhiaiyer\u002Fstatus\u002F2013782002332283180) [ Zach Dive Zach Dive Co-founder & CEO at Adam if you're not using @superset\\_sh, you're getting left behind in 2026 ](https:\u002F\u002Fx.com\u002Fzachdive\u002Fstatus\u002F2014038312508424597) [ Chase McDougall Chase McDougall Founding Engineer at Decoda Health @superset\\_sh is a sick product - love OS since I don't have to wait for someone else to fix bugs ](https:\u002F\u002Fx.com\u002FChaseMcDou\u002Fstatus\u002F2013458004977373643) [ Vlad Arbatov Vlad Arbatov Founding Engineer at Loyal superset became my default tools now so keep the great work folks ](https:\u002F\u002Fx.com\u002Fvladzima\u002Fstatus\u002F2032306550073610246) [ Chris Laupama Chris Laupama TS Lead at Webflow Oh snap @superset\\_sh is soooooo damn good! Now THIS is the experience I've been wanting for CLI agents! ](https:\u002F\u002Fx.com\u002Fchrislaupama\u002Fstatus\u002F2011148329443607037) [ Eric Clemmons Eric Clemmons Principal Engineer at Cloudflare Was using Warp, but now @superset\\_sh has become my primary terminal ](https:\u002F\u002Fx.com\u002Fericclemmons\u002Fstatus\u002F2013413118467056004) [ Leo Leo Co-founder & CTO at Outlit hasn't been a day i haven't used superset since onboarding complete paradigm shift on how i use ai to code ](https:\u002F\u002Fx.com\u002FLeosReal\u002Fstatus\u002F2027306293858586745) [ Elias Ståvik Elias Ståvik Founder at Cleanroom just started using remote desktop instead of ssh just to be able to use @superset\\_sh on my mac mini from my macbook pro generational product in the making, mark my words ](https:\u002F\u002Fx.com\u002Feliasstravik\u002Fstatus\u002F2020580091449708978) [ Gregor Zunic Gregor Zunic Co-founder & CTO at Browser Use Damn @superset\\_sh is really cool. You should try it. worktrees are a breeze cmd + t auto opens Claude Code you can view git changes within itself closing a laptop doesn't kill the sessions ](https:\u002F\u002Fx.com\u002Fgregpr07\u002Fstatus\u002F2013038355630432742) [ Felipe Coury Felipe Coury Codex at OpenAI If you prefer a more GUI-oriented approach to multiple agents in parallel, it seems like @superset\\_sh is doing a tremendous job. ](https:\u002F\u002Fx.com\u002Ffcoury\u002Fstatus\u002F2010477904472281220) [ Iven Iven Engineer at Paraflow Tested various GUI tools for git worktree and agents - Conductor, Vibe Kanban, Agentastic, Crystal, FleetCode, Emdash, Sculptor - but Superset suits my taste the best TranslatedShow original](https:\u002F\u002Fx.com\u002Fivenvd\u002Fstatus\u002F2011738469610242559) Frequently asked questions I already use an IDE like Cursor, is this for me? Which AI coding agents are supported? How does the parallel agent system work? Is Superset free to use? Can I use my own API keys? Try Superset now. Join Waitlist. A Clean \u002F Minimalist, Dark Mode, Typographic \u002F Big Type website in the SaaS & Software industry. The overall color palette features Black, White, Gray. The typography features IBM Plex Mono (Sans Serif), Inter (Sans Serif). Built using Next.js, Node.js, React, Vercel. AI description: Superset is a high-performance developer tool designed to orchestrate and run multiple AI coding agents in parallel. Positioned as \"The Code Editor for AI Agents,\" the platform aims to maximize developer productivity by allowing users to spin up parallel branches of coding tasks, effectively eliminating the downtime typically spent waiting for single-agent execution. It is built for the next generation of software engineers who integrate autonomous agents into their local development workflows. The visual identity is unapologetically technical and modern, utilizing a sophisticated dark mode aesthetic that resonates with the developer community. The design employs heavy, impactful typography and a structured grid layout to showcase social proof from industry giants like Microsoft, OpenAI, and Netflix. By combining a minimalist color palette with high-contrast elements and a clean, functional interface, Superset establishes a sense of power, efficiency, and cutting-edge technological authority.",{"id":473,"website_id":474,"page_id":473,"name":475,"slug":476,"url":477,"website_name":475,"website_slug":476,"website_url":478,"result_url":477,"fetched_at":479,"score":198,"score_boost":346,"ai_score":197,"freshness_score":348,"scored_at":480,"page_name":40,"page_url":477,"page_title":481,"page_description":482,"page_content":483,"page_sort":346,"is_home":353,"is_home_rank":354,"page_type_id":355,"page_type_name":47,"cover":484,"cover_mobile":487,"cover_sequence":489,"translations":580,"industry":21,"styles":583,"credits":584,"font_families":585,"technologies":587,"pagespeed":588,"pagespeed_mobile":593,"buckets":599,"search_payload":601},"ab6e4d8a-c8d8-438c-af32-a8303264e3fe","0299a92c-cc62-40e4-93fe-88dc715ece57","Air","air-dev","https:\u002F\u002Fair.dev\u002F","https:\u002F\u002Fair.dev","2026-05-03T21:18:00.000Z","2026-05-23T19:24:34.000Z","Air: Multitask with agents, stay in control","JetBrains Air is the Agentic Development Environment where Codex, Claude Agent, Gemini CLI, and Junie execute independent task loops without interfering with each other.","Run beloved agents side by side JetBrains Air is the Agentic Development Environment where Codex, Claude Agent, Gemini CLI, and Junie execute independent task loops without interfering with each other. Download for macOS ⤓ PREVIEW WINDOWS AND LINUX ARE COMING HIGHER VELOCITY Run multiple agent tasks in parallel rather than sequentially. BETTER DELEGATION Explore the project, break work apart and assign different agents, matching the right tool to each job. MORE OVERSIGHT Enjoy enhanced visibility into what agents are doing, check in on progress, and course-correct when needed. SMARTER REVIEWS Improve review workflows and test different agents, tools, and flows in one place. HIGHER VELOCITY Run multiple agent tasks in parallel rather than sequentially, getting more done. BETTER DELEGATION Explore the project, break work apart and assign different agents, matching the right tool to each job. MORE OVERSIGHT Enjoy enhanced visibility into what agents are doing, check in on progress, and course-correct when needed. Oversight SMARTER REVIEWS Improve review workflows and test different agents, tools, and flows in one place. a complete loop for coding with agents No more one-shot guessing DEFINE YOUR TASK Explore your project and gather the right context by referencing files, commits, symbols, or images. Define new task interface showing task input and file selectionDefine new task interface showing task input and file selection Run tasks in parallel – without conflicts Launch multiple agents at once and decide where each runs. Choose between Codex, Claude Agent, Gemini CLI, or Junie. Air takes care of the setup with Docker, Git worktrees, or cloud environments (coming soon) – keeping every task isolated. Docker container selection showing Local Workspace, Git Worktree, and Docker optionsDocker container selection showing Local Workspace, Git Worktree, and Docker options Switch context without losing control Get an overview of the agents' progress on your async tasks. Quickly switch to the task to provide additional input or check what the agent is doing. Task list showing multiple tasks running in parallel with progress indicatorsTask list showing multiple tasks running in parallel with progress indicators Review and commit changes Explore your project using language-aware navigation and give agents the right context by referencing files, commits, symbols, or images. Code review interface showing AI suggestions and feedback on changesCode review interface showing AI suggestions and feedback on changes what's coming CLOUD AND AUTOMATIONS Run agents in the cloud without local setup, scaling effortlessly as your needs grow. Start and review tasks from your browser or access cloud agents directly within your JetBrains IDE. Collaborate with your team through shared visibility, automate repetitive workflows, and connect your favorite tools through seamless external integrations. Cloud-native workflowsCloud-native workflows mobile using air at jetbrains With Air, in one evening, I was able to code a pretty sophisticated CLI in Go, a language I had never used before. Vladislav Tankov Director of AI I've been experimenting with Air by adding Junie support to OpenClaw. You get agents like Claude Agent doing the work, but it never feels like you're just watching. Jan-Niklas Wortmann Developer Advocate When I use Air to work on Kotlin microservices, I can easily plan and refine my work while my favorite agents execute in parallel in Git worktrees. Simon Vergauwen Developer Advocate faq +What is Air? Air is an agentic development environment built by JetBrains. With Air, you can delegate coding tasks to multiple AI agents and run them concurrently. It brings your favorite coding agents, Codex, Claude Agent, Gemini CLI, and Junie, into one coherent workflow designed for real codebases. Air helps you define tasks precisely, run them in isolation, and review results with full code intelligence. All in one place. +How is Air different from CLI or chat agents? Most agent tools focus on the model. Air focuses on the workflow around it. Instead of juggling terminals, pasting file paths, and reviewing raw diffs, Air provides: • Code-aware task definition (symbols, classes, methods, commits) • Isolated execution (Docker containers or Git worktrees) • Context-aware review throughout your project • Workspaces to manage concurrent tasks +Which agents are supported? At launch, Air supports: • Claude Agent (Anthropic) • Codex (OpenAI) • Gemini CLI (Google) • Junie (JetBrains) +Can I switch between agents? Yes. Air is agent-agnostic. You can switch models depending on the project or task — or compare outputs side by side. +Can I use my JetBrains AI subscription? JetBrains AI Pro or Ultimate: Sign in and use all supported agents at no additional cost. Please note that JetBrains AI Free and JetBrains AI Enterprise are not supported by Air, and trials cannot be activated from Air. BYOK (Bring Your Own Key): Connect your API keys from Anthropic, OpenAI, or Google. You pay those providers directly at their standard rates. You can also bring your own subscription for Codex and Gemini. You can use both methods (JetBrains AI and BYOK) at the same time. BYOK takes priority when configured, otherwise Air will default to your JetBrains AI subscription. +What are workspaces? Workspaces keep agent sessions, Git state, navigation, and tools scoped to a single project. They let you: • Run tasks concurrently • Preserve session history • Avoid branch collisions • Manage complex workflows without juggling windows +How does Air relate to JetBrains IDEs? Air is a standalone desktop application designed for agent-powered workflows. Air focuses on agent orchestration without replacing existing development workflows. Air handles the agent-powered development; your IDE handles the rest. come get some air Download for macOS ⤓ PREVIEW WINDOWS AND LINUX ARE COMING → Documentation → Follow us on X © 2026 JetBrains. All rights reserved. © 2026 JetBrains. All rights reserved.",{"id":485,"height":13,"width":12,"blurhash":486},"3129e52e-9a86-4fdd-bdd8-396f4e2cae74","radial-gradient(at 0 0,#3b6c78,#00000000 50%),radial-gradient(at 33% 0,#386875,#00000000 50%),radial-gradient(at 67% 0,#306776,#00000000 50%),radial-gradient(at 100% 0,#43707c,#00000000 50%),radial-gradient(at 0 50%,#514c4b,#00000000 50%),radial-gradient(at 33% 50%,#474141,#00000000 50%),radial-gradient(at 67% 50%,#302328,#00000000 50%),radial-gradient(at 100% 50%,#2e1f1f,#00000000 50%),radial-gradient(at 0 100%,#384244,#00000000 50%),radial-gradient(at 33% 100%,#2a3538,#00000000 50%),radial-gradient(at 67% 100%,#21303a,#00000000 50%),radial-gradient(at 100% 100%,#2d3a40,#00000000 50%)",{"id":488,"height":15,"width":14,"blurhash":54},"5705b838-82b3-4871-9e70-4401f331f8f9",[490,493,496,499,502,505,508,511,514,517,520,523,526,529,532,535,538,541,544,547,550,553,556,559,562,565,568,571,574,577],{"directus_files_id":491},{"id":492},"fdcfbb69-6457-4fc7-8cb6-e14e7bfe9d68",{"directus_files_id":494},{"id":495},"ffda39a8-a6f2-47a6-a01b-128c40616568",{"directus_files_id":497},{"id":498},"90ff06df-3272-4185-ab1d-44e3f7635988",{"directus_files_id":500},{"id":501},"98afbf69-0b1c-4440-a281-b818d07b616f",{"directus_files_id":503},{"id":504},"7c6e9f31-b45c-4027-a25f-10464be1997a",{"directus_files_id":506},{"id":507},"5c700a57-167d-4405-a61e-0b17b3a1a64d",{"directus_files_id":509},{"id":510},"828490b6-39d4-4fe2-8f6d-3547dabf6067",{"directus_files_id":512},{"id":513},"14458860-01f5-45b8-9fa9-dc26c8bb8d87",{"directus_files_id":515},{"id":516},"361178a1-6a36-446d-aa27-002560c08c44",{"directus_files_id":518},{"id":519},"0bba2bd0-b6ed-4352-9859-c1135739bad4",{"directus_files_id":521},{"id":522},"0d5ff28e-11c6-4e44-92f0-b564c3eb01a0",{"directus_files_id":524},{"id":525},"7fd36b4d-41e8-4c99-9fc7-57ef36be258a",{"directus_files_id":527},{"id":528},"e2abb8c1-0414-426d-b54b-93731f1f21d2",{"directus_files_id":530},{"id":531},"6e3c8350-3acc-4873-8080-20931ddc128b",{"directus_files_id":533},{"id":534},"a985f5fe-db3f-44b4-bf3c-fd5b20c11136",{"directus_files_id":536},{"id":537},"c92a80c5-19e8-4e40-b8e5-b9353e1c3b22",{"directus_files_id":539},{"id":540},"9ed7d892-f91c-4677-8770-a74ce5ea451e",{"directus_files_id":542},{"id":543},"754a1990-4f88-4720-a73e-5433230295f2",{"directus_files_id":545},{"id":546},"52c6e5ee-732b-4ec8-a1d0-8b99f71668f9",{"directus_files_id":548},{"id":549},"7e51f0eb-6180-4910-9fb8-6fe9a456f770",{"directus_files_id":551},{"id":552},"d110fa44-1034-477f-9f47-c79f7b93752c",{"directus_files_id":554},{"id":555},"75eab981-69ac-4efd-b010-5a31c67053f2",{"directus_files_id":557},{"id":558},"e08ce8fb-6996-4298-9a24-f17c1caa1211",{"directus_files_id":560},{"id":561},"6569c3de-764a-42da-a060-1511467a1b12",{"directus_files_id":563},{"id":564},"3de11420-968b-410e-9339-7ac8d82a9b53",{"directus_files_id":566},{"id":567},"86d60468-3715-410b-bf2a-ff39fd920289",{"directus_files_id":569},{"id":570},"3044e228-4e7c-406e-bbae-81aa6ed13ceb",{"directus_files_id":572},{"id":573},"458c3fee-df89-4f9b-b45a-75e33e01bfd8",{"directus_files_id":575},{"id":576},"7f3a30d6-d001-48dd-9697-d476a2159180",{"directus_files_id":578},{"id":579},"5229e1b6-0a73-4cbc-8613-fdaae399f5b6",[581],{"languages_code":451,"description":582},"Air is a cutting-edge Agentic Development Environment designed to empower developers to orchestrate multiple AI agents—such as Claude, Codex, and Gemini—simultaneously without operational interference. The platform serves as a sophisticated command center for modern software engineering, allowing users to manage complex, independent task loops within a unified, high-performance workspace.\n\nThe visual identity is unapologetically technical and developer-centric, utilizing a sleek dark mode aesthetic paired with high-contrast typography. The design language leans into a 'terminal-chic' vibe, using monospaced fonts and structured UI components that mirror the actual coding environment. It targets high-level engineers and AI researchers who value efficiency, precision, and a toolset that feels as advanced as the agents they are deploying.",[30,252,454],[],[586,296,586],"JetBrains Mono",[],[589,590,591,592],{"score":214,"category":90},{"score":97,"category":93},{"score":97,"category":96},{"score":195,"category":99},[594,595,597,598],{"score":212,"category":90},{"score":596,"category":93},96,{"score":596,"category":96},{"score":195,"category":99},[116,112,600,108],"teal","Website: Air. Page: Homepage. Page type: Home \u002F Landing Page. Page title: Air: Multitask with agents, stay in control. Page description: JetBrains Air is the Agentic Development Environment where Codex, Claude Agent, Gemini CLI, and Junie execute independent task loops without interfering with each other.. Page content: Run beloved agents side by side JetBrains Air is the Agentic Development Environment where Codex, Claude Agent, Gemini CLI, and Junie execute independent task loops without interfering with each other. Download for macOS ⤓ PREVIEW WINDOWS AND LINUX ARE COMING HIGHER VELOCITY Run multiple agent tasks in parallel rather than sequentially. BETTER DELEGATION Explore the project, break work apart and assign different agents, matching the right tool to each job. MORE OVERSIGHT Enjoy enhanced visibility into what agents are doing, check in on progress, and course-correct when needed. SMARTER REVIEWS Improve review workflows and test different agents, tools, and flows in one place. HIGHER VELOCITY Run multiple agent tasks in parallel rather than sequentially, getting more done. BETTER DELEGATION Explore the project, break work apart and assign different agents, matching the right tool to each job. MORE OVERSIGHT Enjoy enhanced visibility into what agents are doing, check in on progress, and course-correct when needed. Oversight SMARTER REVIEWS Improve review workflows and test different agents, tools, and flows in one place. a complete loop for coding with agents No more one-shot guessing DEFINE YOUR TASK Explore your project and gather the right context by referencing files, commits, symbols, or images. Define new task interface showing task input and file selectionDefine new task interface showing task input and file selection Run tasks in parallel – without conflicts Launch multiple agents at once and decide where each runs. Choose between Codex, Claude Agent, Gemini CLI, or Junie. Air takes care of the setup with Docker, Git worktrees, or cloud environments (coming soon) – keeping every task isolated. Docker container selection showing Local Workspace, Git Worktree, and Docker optionsDocker container selection showing Local Workspace, Git Worktree, and Docker options Switch context without losing control Get an overview of the agents' progress on your async tasks. Quickly switch to the task to provide additional input or check what the agent is doing. Task list showing multiple tasks running in parallel with progress indicatorsTask list showing multiple tasks running in parallel with progress indicators Review and commit changes Explore your project using language-aware navigation and give agents the right context by referencing files, commits, symbols, or images. Code review interface showing AI suggestions and feedback on changesCode review interface showing AI suggestions and feedback on changes what's coming CLOUD AND AUTOMATIONS Run agents in the cloud without local setup, scaling effortlessly as your needs grow. Start and review tasks from your browser or access cloud agents directly within your JetBrains IDE. Collaborate with your team through shared visibility, automate repetitive workflows, and connect your favorite tools through seamless external integrations. Cloud-native workflowsCloud-native workflows mobile using air at jetbrains With Air, in one evening, I was able to code a pretty sophisticated CLI in Go, a language I had never used before. Vladislav Tankov Director of AI I've been experimenting with Air by adding Junie support to OpenClaw. You get agents like Claude Agent doing the work, but it never feels like you're just watching. Jan-Niklas Wortmann Developer Advocate When I use Air to work on Kotlin microservices, I can easily plan and refine my work while my favorite agents execute in parallel in Git worktrees. Simon Vergauwen Developer Advocate faq +What is Air? Air is an agentic development environment built by JetBrains. With Air, you can delegate coding tasks to multiple AI agents and run them concurrently. It brings your favorite coding agents, Codex, Claude Agent, Gemini CLI, and Junie, into one coherent workflow designed for real codebases. Air helps you define tasks precisely, run them in isolation, and review results with full code intelligence. All in one place. +How is Air different from CLI or chat agents? Most agent tools focus on the model. Air focuses on the workflow around it. Instead of juggling terminals, pasting file paths, and reviewing raw diffs, Air provides: • Code-aware task definition (symbols, classes, methods, commits) • Isolated execution (Docker containers or Git worktrees) • Context-aware review throughout your project • Workspaces to manage concurrent tasks +Which agents are supported? At launch, Air supports: • Claude Agent (Anthropic) • Codex (OpenAI) • Gemini CLI (Google) • Junie (JetBrains) +Can I switch between agents? Yes. Air is agent-agnostic. You can switch models depending on the project or task — or compare outputs side by side. +Can I use my JetBrains AI subscription? JetBrains AI Pro or Ultimate: Sign in and use all supported agents at no additional cost. Please note that JetBrains AI Free and JetBrains AI Enterprise are not supported by Air, and trials cannot be activated from Air. BYOK (Bring Your Own Key): Connect your API keys from Anthropic, OpenAI, or Google. You pay those providers directly at their standard rates. You can also bring your own subscription for Codex and Gemini. You can use both methods (JetBrains AI and BYOK) at the same time. BYOK takes priority when configured, otherwise Air will default to your JetBrains AI subscription. +What are workspaces? Workspaces keep agent sessions, Git state, navigation, and tools scoped to a single project. They let you: • Run tasks concurrently • Preserve session history • Avoid branch collisions • Manage complex workflows without juggling windows +How does Air relate to JetBrains IDEs? Air is a standalone desktop application designed for agent-powered workflows. Air focuses on agent orchestration without replacing existing development workflows. Air handles the agent-powered development; your IDE handles the rest. come get some air Download for macOS ⤓ PREVIEW WINDOWS AND LINUX ARE COMING → Documentation → Follow us on X © 2026 JetBrains. All rights reserved. © 2026 JetBrains. All rights reserved.. A Clean \u002F Minimalist, Dark Mode, Typographic \u002F Big Type website in the SaaS & Software industry. The overall color palette features Black, Gray, Teal, White. The typography features JetBrains Mono (Sans Serif), Inter (Sans Serif), JetBrains Mono (Sans Serif). AI description: Air is a cutting-edge Agentic Development Environment designed to empower developers to orchestrate multiple AI agents—such as Claude, Codex, and Gemini—simultaneously without operational interference. The platform serves as a sophisticated command center for modern software engineering, allowing users to manage complex, independent task loops within a unified, high-performance workspace. The visual identity is unapologetically technical and developer-centric, utilizing a sleek dark mode aesthetic paired with high-contrast typography. The design language leans into a 'terminal-chic' vibe, using monospaced fonts and structured UI components that mirror the actual coding environment. It targets high-level engineers and AI researchers who value efficiency, precision, and a toolset that feels as advanced as the agents they are deploying.",{"id":603,"website_id":604,"page_id":603,"name":605,"slug":606,"url":607,"website_name":605,"website_slug":606,"website_url":608,"result_url":607,"fetched_at":609,"score":94,"score_boost":346,"ai_score":347,"freshness_score":610,"scored_at":611,"page_name":40,"page_url":607,"page_title":612,"page_description":613,"page_content":614,"page_sort":346,"is_home":353,"is_home_rank":354,"page_type_id":355,"page_type_name":47,"cover":615,"cover_mobile":618,"cover_sequence":620,"translations":711,"industry":21,"styles":714,"credits":716,"font_families":717,"technologies":719,"pagespeed":720,"pagespeed_mobile":727,"buckets":732,"search_payload":733},"9db83877-7ac5-4dd0-8940-9b2eb0f1e992","e53e3c14-da75-459c-aa66-32a905a96c1f","Greptile","greptile","https:\u002F\u002Fgreptile.com\u002F","https:\u002F\u002Fgreptile.com","2026-05-01T14:09:40.000Z",71,"2026-05-23T19:23:46.000Z","AI Code Review | Greptile | Merge 4X Faster, Catch 3X More Bugs","AI Code Reviews that understand your entire codebase. Automate PR reviews, catch bugs faster, improve code quality with AI-driven analysis. Try Greptile free!","AI Content Index The AI Code Reviewer. AI agents that review and test pull requests with full context of the codebase. Contact SalesStart now Over 9,000+ teams use Greptile Substack Klaviyo Retool Nvidia Brex Scale PostHog Mintlify Zapier Joby WorkOS Compass Bilt Crossmint Browserbase Nominal Brex logo > ### “We've tried more code review tools than I can count. _Greptile outperforms them all_ by a mile. Honestly the only AI reviewer that doesn't annoy the s\\\\t out of me.” James Reggio CTO @ Brex James Reggio View Brex case study \\[ AGENT + How Greptile reviews every PR Greptile constructs a graph index of your codebase, then uses a swarm of agents to catch potential issues that humans might miss Learn more How it works STEP 01 Indexes your codebase Builds a graph of your repo - files, functions, and dependencies. STEP 02 Swarm of agents review the PR Parallel agents review changes, assess their impact beyond the diff, and flag issues. STEP 03 Greptile learns your codebase over time Reads other engineers' comments to understand your coding standards. Catch them all From style violations to security risks and multi-file logical bugs Three bugs that should not hit prod. NVIDIA Guardrails 9.2k Stars 1.5k Forks 3 Repositories guardrails\u002Fllm\\_output.py 01\\- 01+ 02+ Greptilegreptile Reasoning traces leak into responses Netflix Metaflow 8.5k Stars 1.3k Forks netflix\u002Fmetaflow Repository metaflow\u002Fclient\u002Fapi.py 01\\- 01+ 02+ Greptilegreptile Double retry causes up to N² attempts Meta PyTorch 1.1k Stars 176 Forks meta-pytorch\u002FOpenEnv Repository openenv\u002Fsandbox.py 01\\- 01+ 02+ Greptilegreptile Sandbox escape via Python introspection builtins See more examples \\[ PERSONALIZATION \\] Reviews that adapt to your team From custom rules to continuous learning, Greptile gets smarter about your codebase with every review. Custom rules \\[ CUSTOM RULES \\] Your house, your rules Write standards in plain English, point Greptile at repo-specific context, and enforce the patterns your team actually cares about on every PR. Set your rules \\[ LEARNING \\] We're getting to know each other Greptile learns your codebase and coding standards by reading your team' PR comments. See learning \\[ YOUR STACK \\] The Central Validation Layer. Designed to work seamlessly with every coding agent, Greptile serves as the unified validation agent for all code changes. Learn more Ways Greptile plugs into your workflow Fix in your IDE 01 Fix in your IDE One-click send issue context to Claude Code, Cursor, Codex, or Devin Greptile MCP 02 Greptile MCP Connect Greptile to any AI agent to share comment context. Claude Code Plugin 03 Claude Code Plugin Let Claude Code automatically read and resolve Greptile comments. \u002Fgreploop 04 \u002Fgreploop Let any coding agent iterate with Greptile until all issues are resolved. TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE \\[ INTRODUCING TREX \\] Autonomously Test Every PR TREX is a Greptile agent that writes and runs tests for every PR in a sandbox to catch bugs and missed edge cases. Get early access TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE Security-First Design Built for enterprises across defense, healthcare, and financial services. Learn more 01 Self-Hosted Deployment Host Greptile in your own air-gapped environment. 02 SOC 2 Compliance Independent audits, reports available on request. 03 Enterprise Ready SSO, audit logs, dedicated support. James James ReggioCTO · Brex Brex logo > We've tried more code review tools than I can count. Greptile outperforms them all by a mile. Honestly the only AI reviewer that doesn't annoy the s\\\\t out of me. Mark Mark TranEng. Manager · WorkOS WorkOS logo > We've been impressed by Greptile's code review quality. It's tightened our feedback loops, improved consistency, and freed up engineers to focus on higher-level design and architecture. See what our customers are saying View testimonials Anirudh Anirudh KamathTech Lead · Browserbase Browserbase logo > Greptile helps the team do their best work. It levels everybody up and helps ensure that we're all proud of our code Hahnbee Hahnbee LeeCTO · Mintlify Mintlify logo > I'm thankful for the things Greptile catches in my pull requests. Isaiah Isaiah GranetCEO · Bland.com Bland.com logo > Has been a game changer. Couldn't ever go back.... If you are on the fence — just try it. Chase Chase GladishCTO · RollCredits RollCredits logo > Greptile has been the first AI code review tool I've used that actually provides useful feedback on PRs. It's catching issues that human reviewers easily miss and giving senior-level suggestions to boot FAQ How does Greptile pricing work? $30\u002Fseat\u002Fmonth. 50 reviews included per seat. $1 per additional review. Can Greptile be self-hosted? Yes, you can self-host Greptile in your AWS environment and even use your own LLM providers for added flexibility. Are there free trials or discounts available for Greptile? Greptile offers free usage for qualified open source projects and significant discounts for pre-Series A startups. Contact us to check your eligibility. What programming languages does Greptile support? Greptile fully supports Python, JavaScript, TypeScript, Go, Elixir, Java, C, C++, C#, Swift, PHP, and Rust. Most other languages are supported with slightly lower response quality. Is Greptile compatible with GitLab? Yes, Greptile integrates seamlessly with GitLab in addition to GitHub. Can I use Greptile's API for my own product? Yes, we offer discounted bulk pricing for API use. Please contact us at support@greptile.com for more information. * What is AI code review? AI code review learns your codebase and applies repository context to analyze pull requests, catching syntax, logic, and style issues and suggesting fixes. Learn more in our AI code review guide. Can't find your answer here? Get in touch. Greptile is building the code validation layer so you can get back to shipping. View PricingStart now",{"id":616,"height":13,"width":12,"blurhash":617},"4c96bac4-a490-4435-9f6a-be4d6a01e5da","radial-gradient(at 0 0,#e6e7e4,#00000000 50%),radial-gradient(at 33% 0,#e0e3e0,#00000000 50%),radial-gradient(at 67% 0,#e9ebe9,#00000000 50%),radial-gradient(at 100% 0,#e8ebe8,#00000000 50%),radial-gradient(at 0 50%,#d2d1d3,#00000000 50%),radial-gradient(at 33% 50%,#d0cfd0,#00000000 50%),radial-gradient(at 67% 50%,#d8d8d8,#00000000 50%),radial-gradient(at 100% 50%,#dad9da,#00000000 50%),radial-gradient(at 0 100%,#dce0dd,#00000000 50%),radial-gradient(at 33% 100%,#dde0de,#00000000 50%),radial-gradient(at 67% 100%,#e5e6e5,#00000000 50%),radial-gradient(at 100% 100%,#dcdddc,#00000000 50%)",{"id":619,"height":15,"width":14,"blurhash":54},"9cb1307d-f1c9-4977-a913-f65fe51b0064",[621,624,627,630,633,636,639,642,645,648,651,654,657,660,663,666,669,672,675,678,681,684,687,690,693,696,699,702,705,708],{"directus_files_id":622},{"id":623},"b7cbbc35-fc1b-40cf-bb77-34ed1638045b",{"directus_files_id":625},{"id":626},"1f5f2bce-368e-417f-b42f-d09b2d8a163d",{"directus_files_id":628},{"id":629},"da1744f2-84e6-472f-9a64-a854b2d16448",{"directus_files_id":631},{"id":632},"661494e9-bd72-4327-b25c-8c2521967d84",{"directus_files_id":634},{"id":635},"bc2b27bb-1737-4929-92bd-da57c05c0421",{"directus_files_id":637},{"id":638},"5b8eaf2b-8eda-45f7-b161-56e88f1247e8",{"directus_files_id":640},{"id":641},"a2b426b0-d0a3-4edb-ba66-b69039b90a10",{"directus_files_id":643},{"id":644},"4fe4458b-5598-44c2-8b84-70b48cd7ead2",{"directus_files_id":646},{"id":647},"23206236-5edb-4efc-97dc-79380859f6ce",{"directus_files_id":649},{"id":650},"c9009a59-b903-4abc-bfaf-6368310ae5d9",{"directus_files_id":652},{"id":653},"6f570ba2-9f58-409f-8118-ef1ec0bb6850",{"directus_files_id":655},{"id":656},"f77c7268-c23b-4007-8828-27b83698aacf",{"directus_files_id":658},{"id":659},"3b0789b4-611f-4871-811b-e44a83cdf6d7",{"directus_files_id":661},{"id":662},"8b5f17fd-9aa1-49a0-b00c-4db28ac63599",{"directus_files_id":664},{"id":665},"052c9fe0-fa8d-4aed-aaac-ab7b352208d4",{"directus_files_id":667},{"id":668},"649701d0-ff27-47b4-b273-8115e8df4b07",{"directus_files_id":670},{"id":671},"4e66792a-c55f-4b94-926a-3a4e2b80c137",{"directus_files_id":673},{"id":674},"6fecf7fc-6210-4069-8755-8a28f402c97a",{"directus_files_id":676},{"id":677},"927c8d72-215f-404a-b58f-05e070d1a357",{"directus_files_id":679},{"id":680},"ac7ac9c3-2b0a-4e87-bca4-d118d36e2fbe",{"directus_files_id":682},{"id":683},"8299d5f8-78f7-4a7a-82c1-b084a7bd1c35",{"directus_files_id":685},{"id":686},"44a6a377-2d90-47c4-b313-28997ab7208f",{"directus_files_id":688},{"id":689},"45befd80-e6cb-4ace-8fef-4e99d8c99413",{"directus_files_id":691},{"id":692},"d81eada7-58aa-4c92-8699-547e3f7ddb49",{"directus_files_id":694},{"id":695},"17f894a0-5352-48b1-a6e2-ab578425cc2d",{"directus_files_id":697},{"id":698},"bbaf8dd4-9a9e-4c3e-9528-b5929b3b6b83",{"directus_files_id":700},{"id":701},"0df37127-cb9a-4ea4-9daa-0daffe13ee77",{"directus_files_id":703},{"id":704},"b4e2612c-e1f0-4109-8bcd-75690c8c9699",{"directus_files_id":706},{"id":707},"15f538b3-8c6c-4c62-9603-1423b57d45bb",{"directus_files_id":709},{"id":710},"10d9811a-b639-4b60-8e38-f3cd7c9b6100",[712],{"languages_code":451,"description":713},"Greptile is a sophisticated AI-driven developer tool designed to automate code reviews and pull request testing by understanding the full context of a codebase. By deploying intelligent AI agents, the platform aims to accelerate engineering velocity, allowing teams to merge code significantly faster while simultaneously reducing bug density. It targets high-growth engineering organizations and DevOps teams looking to scale their quality assurance processes without increasing manual overhead.\n\nThe visual identity is striking and unconventional for the B2B SaaS sector, leaning heavily into a high-contrast, neo-brutalist aesthetic. The design utilizes a muted, grid-based layout paired with bold, oversized typography and a distinctive halftone-style graphic element. A vibrant mint green accent provides a sharp pop of color against a sophisticated grayscale palette, creating a look that feels both cutting-edge and intellectually rigorous, perfectly mirroring the advanced nature of its underlying technology.",[715,30,252],"Brutalist \u002F Neo-Brutalist",[],[315,718],"Space Mono",[],[721,723,724,726],{"score":722,"category":90},43,{"score":91,"category":93},{"score":725,"category":96},77,{"score":195,"category":99},[728,729,730,731],{"score":227,"category":90},{"score":91,"category":93},{"score":725,"category":96},{"score":195,"category":99},[108,112,600],"Website: Greptile. Page: Homepage. Page type: Home \u002F Landing Page. Page title: AI Code Review | Greptile | Merge 4X Faster, Catch 3X More Bugs. Page description: AI Code Reviews that understand your entire codebase. Automate PR reviews, catch bugs faster, improve code quality with AI-driven analysis. Try Greptile free!. Page content: AI Content Index The AI Code Reviewer. AI agents that review and test pull requests with full context of the codebase. Contact SalesStart now Over 9,000+ teams use Greptile Substack Klaviyo Retool Nvidia Brex Scale PostHog Mintlify Zapier Joby WorkOS Compass Bilt Crossmint Browserbase Nominal Brex logo > ### “We've tried more code review tools than I can count. _Greptile outperforms them all_ by a mile. Honestly the only AI reviewer that doesn't annoy the s\\\\t out of me.” James Reggio CTO @ Brex James Reggio View Brex case study \\[ AGENT + How Greptile reviews every PR Greptile constructs a graph index of your codebase, then uses a swarm of agents to catch potential issues that humans might miss Learn more How it works STEP 01 Indexes your codebase Builds a graph of your repo - files, functions, and dependencies. STEP 02 Swarm of agents review the PR Parallel agents review changes, assess their impact beyond the diff, and flag issues. STEP 03 Greptile learns your codebase over time Reads other engineers' comments to understand your coding standards. Catch them all From style violations to security risks and multi-file logical bugs Three bugs that should not hit prod. NVIDIA Guardrails 9.2k Stars 1.5k Forks 3 Repositories guardrails\u002Fllm\\_output.py 01\\- 01+ 02+ Greptilegreptile Reasoning traces leak into responses Netflix Metaflow 8.5k Stars 1.3k Forks netflix\u002Fmetaflow Repository metaflow\u002Fclient\u002Fapi.py 01\\- 01+ 02+ Greptilegreptile Double retry causes up to N² attempts Meta PyTorch 1.1k Stars 176 Forks meta-pytorch\u002FOpenEnv Repository openenv\u002Fsandbox.py 01\\- 01+ 02+ Greptilegreptile Sandbox escape via Python introspection builtins See more examples \\[ PERSONALIZATION \\] Reviews that adapt to your team From custom rules to continuous learning, Greptile gets smarter about your codebase with every review. Custom rules \\[ CUSTOM RULES \\] Your house, your rules Write standards in plain English, point Greptile at repo-specific context, and enforce the patterns your team actually cares about on every PR. Set your rules \\[ LEARNING \\] We're getting to know each other Greptile learns your codebase and coding standards by reading your team' PR comments. See learning \\[ YOUR STACK \\] The Central Validation Layer. Designed to work seamlessly with every coding agent, Greptile serves as the unified validation agent for all code changes. Learn more Ways Greptile plugs into your workflow Fix in your IDE 01 Fix in your IDE One-click send issue context to Claude Code, Cursor, Codex, or Devin Greptile MCP 02 Greptile MCP Connect Greptile to any AI agent to share comment context. Claude Code Plugin 03 Claude Code Plugin Let Claude Code automatically read and resolve Greptile comments. \u002Fgreploop 04 \u002Fgreploop Let any coding agent iterate with Greptile until all issues are resolved. TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE \\[ INTRODUCING TREX \\] Autonomously Test Every PR TREX is a Greptile agent that writes and runs tests for every PR in a sandbox to catch bugs and missed edge cases. Get early access TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE TEST · RUN · EXECUTE Security-First Design Built for enterprises across defense, healthcare, and financial services. Learn more 01 Self-Hosted Deployment Host Greptile in your own air-gapped environment. 02 SOC 2 Compliance Independent audits, reports available on request. 03 Enterprise Ready SSO, audit logs, dedicated support. James James ReggioCTO · Brex Brex logo > We've tried more code review tools than I can count. Greptile outperforms them all by a mile. Honestly the only AI reviewer that doesn't annoy the s\\\\t out of me. Mark Mark TranEng. Manager · WorkOS WorkOS logo > We've been impressed by Greptile's code review quality. It's tightened our feedback loops, improved consistency, and freed up engineers to focus on higher-level design and architecture. See what our customers are saying View testimonials Anirudh Anirudh KamathTech Lead · Browserbase Browserbase logo > Greptile helps the team do their best work. It levels everybody up and helps ensure that we're all proud of our code Hahnbee Hahnbee LeeCTO · Mintlify Mintlify logo > I'm thankful for the things Greptile catches in my pull requests. Isaiah Isaiah GranetCEO · Bland.com Bland.com logo > Has been a game changer. Couldn't ever go back.... If you are on the fence — just try it. Chase Chase GladishCTO · RollCredits RollCredits logo > Greptile has been the first AI code review tool I've used that actually provides useful feedback on PRs. It's catching issues that human reviewers easily miss and giving senior-level suggestions to boot FAQ How does Greptile pricing work? $30\u002Fseat\u002Fmonth. 50 reviews included per seat. $1 per additional review. Can Greptile be self-hosted? Yes, you can self-host Greptile in your AWS environment and even use your own LLM providers for added flexibility. Are there free trials or discounts available for Greptile? Greptile offers free usage for qualified open source projects and significant discounts for pre-Series A startups. Contact us to check your eligibility. What programming languages does Greptile support? Greptile fully supports Python, JavaScript, TypeScript, Go, Elixir, Java, C, C++, C#, Swift, PHP, and Rust. Most other languages are supported with slightly lower response quality. Is Greptile compatible with GitLab? Yes, Greptile integrates seamlessly with GitLab in addition to GitHub. Can I use Greptile's API for my own product? Yes, we offer discounted bulk pricing for API use. Please contact us at support@greptile.com for more information. * What is AI code review? AI code review learns your codebase and applies repository context to analyze pull requests, catching syntax, logic, and style issues and suggesting fixes. Learn more in our AI code review guide. Can't find your answer here? Get in touch. Greptile is building the code validation layer so you can get back to shipping. View PricingStart now. A Brutalist \u002F Neo-Brutalist, Clean \u002F Minimalist, Dark Mode website in the SaaS & Software industry. The overall color palette features White, Gray, Teal. The typography features DM Sans (Sans Serif), Space Mono (Colophon). AI description: Greptile is a sophisticated AI-driven developer tool designed to automate code reviews and pull request testing by understanding the full context of a codebase. By deploying intelligent AI agents, the platform aims to accelerate engineering velocity, allowing teams to merge code significantly faster while simultaneously reducing bug density. It targets high-growth engineering organizations and DevOps teams looking to scale their quality assurance processes without increasing manual overhead. The visual identity is striking and unconventional for the B2B SaaS sector, leaning heavily into a high-contrast, neo-brutalist aesthetic. The design utilizes a muted, grid-based layout paired with bold, oversized typography and a distinctive halftone-style graphic element. A vibrant mint green accent provides a sharp pop of color against a sophisticated grayscale palette, creating a look that feels both cutting-edge and intellectually rigorous, perfectly mirroring the advanced nature of its underlying technology.",{"id":735,"website_id":736,"page_id":735,"name":737,"slug":738,"url":739,"website_name":737,"website_slug":738,"website_url":740,"result_url":739,"fetched_at":741,"score":198,"score_boost":346,"ai_score":347,"freshness_score":153,"scored_at":742,"page_name":40,"page_url":739,"page_title":743,"page_description":744,"page_content":745,"page_sort":346,"is_home":353,"is_home_rank":354,"page_type_id":355,"page_type_name":47,"cover":746,"cover_mobile":749,"cover_sequence":751,"translations":842,"industry":21,"styles":845,"credits":847,"font_families":848,"technologies":849,"pagespeed":850,"pagespeed_mobile":855,"buckets":860,"search_payload":862},"60bb0bb8-52be-4fd8-8023-9f134e39444d","dc0566d6-00bf-49bb-aa33-96bb0a8a7977","Steel","steel","https:\u002F\u002Fsteel.dev\u002F","https:\u002F\u002Fsteel.dev","2026-05-05T15:01:02.000Z","2026-05-23T19:22:03.000Z","Steel | Open-source Headless Browser API","Steel is an open-source browser API purpose-built for AI agents.","If you are an AI agent, read install.md for step-by-step setup and usage instructions. Also see https:\u002F\u002Fdocs.steel.dev\u002Fllms.txt for more information. [ ](.\u002F#hero) Sessions API Pricing Blog Docs We're hiring! [ 6.9K ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-browser) [ Log in \u002F Sign Up ](https:\u002F\u002Fapp.steel.dev\u002Fsign-in) Browser Infrastructure for AI Agents Steel is an open source browser API that lets you control fleets of browsers in the cloud. [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) [ Read Docs ](https:\u002F\u002Fdocs.steel.dev\u002F) MY\\_APP Can you book me a flight to Madrid for 2 in November? Searching for flights from AMS to MAD.. Found 12 matching flights, would you prioritize low price or time? Prioritize time over price Found 12 matching flights, would you prioritize low price or time? AI\\_LOGIC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 import Steel from ‘steel\\-sdk’; import OpenAI from 'openai'; const client = new OpenAI({ apiKey: process.env\\['OPENAI\\_API\\_KEY'\\], \u002F\u002F This is the default and can be omitted}); async function main() { const chatCompletion = await client.chat.completions.create({ messages: \\[{ role: 'user', content: 'Say this is a test' }\\], model: 'gpt-3.5-turbo', }); } main(); STEEL\\_BROWSER skyscanner.net 20:05 AMS 13:45 MAD 20:05 AMS 20:05 AMS 20:05 AMS 20:05 AMS 800B+ Tokens Scraped 800,000+ Browser Hours Served \u003C1s Avg. Session Start Time Use Cases What Developers Build on Steel From large-scale scrape jobs to fully autonomous web agents, Steel makes it easy to run browser automations in the cloud. AI Agents Foundational Model Training AI Shopping Assistants Web Scraping RPA Software Sales Automation QA Software Customer Service Agents Sessions API Spin up on-demand browser sessions with a simple API call. [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) I’m not a robot Auto CAPTCHA solving Built-in CAPTCHA solving that keeps your automation flowing. Proxy and Browser Fingerprinting Simple controls to never worry about getting flagged as a bot again. 500 ms Steel 5000 ms Competition Quick-start Average session starts in less than 1s when client is in same region. Live - 23h:59m Up to 24h Long Sessions Run for a minute or several hours, each session can run up to 24 hours. cookie#1 cookie#4 cookie#3 cookie#2 Manage and Reuse Context Save and inject cookies and local storage to pick up where you left off. \\-puppeteer.launch ({...}) +puppeteer.connect ({...}) 1-line Change to Run Puppeteer Easily run your Puppeteer, Playwright, or Selenium in the cloud. Playing World-class Observability Session Viewer lets you view and debug live or recorded sessions. Sign in Securely Auto Sign-In Securely give your agents access to auth-walled websites and apps. Get Started with Your Favorite Stack Easily integrate with the tools you know and love. We're always working on adding more to this list. [ Docs ](https:\u002F\u002Fdocs.steel.dev\u002F) Python Node.js Puppeteer Playwright Selenium Coming Soon from steel import Steel client = Steel() \\# Create a session session = client.sessions.create() print(f\"Session created: {session.id}\") \\# Your session is now ready to use! \\# Keep this ID handy - you'll need it to connect with automation tools [ view Python SDK ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-python) Cookbook Examples Try out Steel with the brand new Cookbook examples [ View All Examples ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook) Browser Use Learn how to build an AI Agent with Steel and browser-use [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-browser-use-starter) Magnitude Learn how to build an AI Agent with Steel and Magnitude [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-magnitude-starter) Notte Learn how to build an AI Agent with Steel and Notte [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-notte-starter) OpenAI's Computer Use Learn how to create a browser automation agent with Steel and OpenAI's Computer Use [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-oai-computer-use-node-starter) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-oai-computer-use-python-starter) Playwright Learn how to build a starter project for using Playwright with Steel [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-playwright-starter-js) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-playwright-python-starter) Claude's Computer Use Learn how to create a browser automation agent with Steel and Claude's Computer use [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-claude-computer-use-node-starter) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-claude-computer-use-python-starter) Our Pricing For info, limits and rates, explore a detailed pricing breakdown ↗. Monthly Annually (Save 20%) Hobby Free $10 in Credits per month Everything you need to start automating browsers. Monthly: 100 Browser Hours [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002F) Starter $29 $29 in Credits per month Everything you need to start automating browsers. Monthly: 290 Browser Hours 2.9 GB Proxy Bandwidth 7.2k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Developers Popular $99 $99 in Credits per month Everything you need to launch your project. Monthly: 1,238 Browser Hours 12 GB Proxy Bandwidth 28k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Startups $499 $499 in Credits per month Everything you need to scale your masterpiece. Monthly: 9,980 Browser Hours 166 GB Proxy Bandwidth 166k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Enterprise Spin up thousands of browser sessions on the cloud. [ Book a Demo ](https:\u002F\u002Fcal.com\u002Fhussien-hussien-fjxt3x\u002Fsteel-enterprise-demo) Custom Pricing Higher Limits Discounted Rates Dedicated Clusters & APIs Unlimited Session Concurrency GitHub Repository Steel is proudly open source. Run our sessions locally or self-hosted with our docker container. [ steel-browser ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-browser) Research Grants If you’re doing research for a school, send us an email from your school emails for grants in the form of free credits. Email us at research@steel.dev Ready to Build with Steel? [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) [ Documentation ](https:\u002F\u002Fdocs.steel.dev\u002F) A better way to take your LLMs online. © Nen Labs · Inc. 2026. Terms of Service · Privacy Policy [ All Systems Operational ](https:\u002F\u002Fstatus.steel.dev\u002F) Platform Docs Status Brand Assets Launch Week @ Steel Join the Community X \u002F Twitter GitHub Blog Discord Careers ![](https:\u002F\u002Fanalytics.twitter.com\u002F1\u002Fi\u002Fadsct?bci=4&dv=UTC%26en-US%26Google%20Inc.%26Linux%20x86_64%26255%26800%26600%264%2624%26800%26600%260%26na&eci=3&event=%7B%7D&event_id=be57c6d2-542b-41e7-8915-59accfbb7df3&integration=advert …",{"id":747,"height":13,"width":12,"blurhash":748},"057633d0-5f9d-4521-9bc5-3d91b7b7a809","radial-gradient(at 0 0,#284553,#00000000 50%),radial-gradient(at 33% 0,#341934,#00000000 50%),radial-gradient(at 67% 0,#6b8798,#00000000 50%),radial-gradient(at 100% 0,#6a98a1,#00000000 50%),radial-gradient(at 0 50%,#3c4859,#00000000 50%),radial-gradient(at 33% 50%,#4b4446,#00000000 50%),radial-gradient(at 67% 50%,#6c8994,#00000000 50%),radial-gradient(at 100% 50%,#66929d,#00000000 50%),radial-gradient(at 0 100%,#293744,#00000000 50%),radial-gradient(at 33% 100%,#380c22,#00000000 50%),radial-gradient(at 67% 100%,#5d818e,#00000000 50%),radial-gradient(at 100% 100%,#5a8e98,#00000000 50%)",{"id":750,"height":15,"width":14,"blurhash":54},"88c0f3ac-b038-446c-9752-d7c9fef5a212",[752,755,758,761,764,767,770,773,776,779,782,785,788,791,794,797,800,803,806,809,812,815,818,821,824,827,830,833,836,839],{"directus_files_id":753},{"id":754},"9570f3d8-b3e7-483a-8982-0d4b7ea6f529",{"directus_files_id":756},{"id":757},"f8754fd0-65f1-417d-b947-305eae1fdba8",{"directus_files_id":759},{"id":760},"e283b8c4-ecd4-4fc9-910a-0e6de31a8de0",{"directus_files_id":762},{"id":763},"7c36a5b4-90e8-4648-af65-8dd24f74d392",{"directus_files_id":765},{"id":766},"84e0218d-b9ff-49ae-a66d-22215730971c",{"directus_files_id":768},{"id":769},"68b77108-f8c0-43e3-82fc-c2124ea7e35c",{"directus_files_id":771},{"id":772},"778d8bc9-f61e-4f73-93fd-3d0a7d049850",{"directus_files_id":774},{"id":775},"55c36a10-3f2b-481e-9926-ddc441b86696",{"directus_files_id":777},{"id":778},"7b82d52d-a773-4b00-bd29-1d191fc74166",{"directus_files_id":780},{"id":781},"dc406009-7da3-4ed2-b410-7131df39ed40",{"directus_files_id":783},{"id":784},"244c5c4d-66c0-43f5-88e3-cb2f92128be8",{"directus_files_id":786},{"id":787},"7e999c71-0971-4dec-b7b4-7bd280f38325",{"directus_files_id":789},{"id":790},"9339006b-892f-4609-ba46-7bef530e79bd",{"directus_files_id":792},{"id":793},"716ffa01-71d2-4ce1-8ffe-cd38046fb84c",{"directus_files_id":795},{"id":796},"93cbf71e-865a-4c06-89ec-bcf8b1a273b5",{"directus_files_id":798},{"id":799},"052d8021-d68e-4ba9-bcab-0124efda8e80",{"directus_files_id":801},{"id":802},"47674690-7fb1-4c40-8a3e-021a227b3931",{"directus_files_id":804},{"id":805},"9ef6aa6e-3f38-4b2b-8b6b-789b738a5b5a",{"directus_files_id":807},{"id":808},"cd2577cc-6ad6-464e-a572-5b13107eb14a",{"directus_files_id":810},{"id":811},"3cc3f31c-172b-4909-9733-e0f59b55d764",{"directus_files_id":813},{"id":814},"823cefed-5ffa-4e83-8e0c-c74532ab52ab",{"directus_files_id":816},{"id":817},"2a68bb2a-5313-4b7a-b67b-74c2f2a493a3",{"directus_files_id":819},{"id":820},"19da94d2-e752-46d0-9c0a-3e7a0292a6fd",{"directus_files_id":822},{"id":823},"f9bd6621-66ef-489a-9d00-40d2441f3f82",{"directus_files_id":825},{"id":826},"c8557cea-bdfe-4480-85a5-a9c76964cfd0",{"directus_files_id":828},{"id":829},"2e67a9bc-3244-45a6-9289-6282dbf69980",{"directus_files_id":831},{"id":832},"4a2ca72e-970f-49d1-a7ed-e850bba4bf73",{"directus_files_id":834},{"id":835},"31028cd2-0de8-4a68-86bd-f7d51c838038",{"directus_files_id":837},{"id":838},"e8dd1eae-6b2c-4024-be58-294b0bf03c8b",{"directus_files_id":840},{"id":841},"f1e953bc-a421-4349-87e5-d6288935da76",[843],{"languages_code":451,"description":844},"Steel provides a high-performance, open-source headless browser API specifically engineered to power the next generation of AI agents. By allowing developers to control vast fleets of browsers in the cloud, the platform bridges the gap between LLM reasoning and real-world web interaction, enabling seamless automation and data scraping at scale.\n\nThe visual identity is unapologetically developer-centric, utilizing a sophisticated dark mode aesthetic paired with high-contrast accents. The design employs a gritty, textured background that evokes a sense of raw, industrial power, while clean typography and structured UI components maintain a sense of technical precision. It is a brand built for engineers who value both speed and robust infrastructure.",[30,252,846],"Futuristic \u002F Sci-Fi",[],[52,56,52,52,296,586],[291,72],[851,852,853,854],{"score":174,"category":90},{"score":209,"category":93},{"score":596,"category":96},{"score":97,"category":99},[856,857,858,859],{"score":211,"category":90},{"score":209,"category":93},{"score":97,"category":96},{"score":97,"category":99},[116,108,600,112,861],"yellow","Website: Steel. Page: Homepage. Page type: Home \u002F Landing Page. Page title: Steel | Open-source Headless Browser API. Page description: Steel is an open-source browser API purpose-built for AI agents.. Page content: If you are an AI agent, read install.md for step-by-step setup and usage instructions. Also see https:\u002F\u002Fdocs.steel.dev\u002Fllms.txt for more information. [ ](.\u002F#hero) Sessions API Pricing Blog Docs We're hiring! [ 6.9K ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-browser) [ Log in \u002F Sign Up ](https:\u002F\u002Fapp.steel.dev\u002Fsign-in) Browser Infrastructure for AI Agents Steel is an open source browser API that lets you control fleets of browsers in the cloud. [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) [ Read Docs ](https:\u002F\u002Fdocs.steel.dev\u002F) MY\\_APP Can you book me a flight to Madrid for 2 in November? Searching for flights from AMS to MAD.. Found 12 matching flights, would you prioritize low price or time? Prioritize time over price Found 12 matching flights, would you prioritize low price or time? AI\\_LOGIC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 import Steel from ‘steel\\-sdk’; import OpenAI from 'openai'; const client = new OpenAI({ apiKey: process.env\\['OPENAI\\_API\\_KEY'\\], \u002F\u002F This is the default and can be omitted}); async function main() { const chatCompletion = await client.chat.completions.create({ messages: \\[{ role: 'user', content: 'Say this is a test' }\\], model: 'gpt-3.5-turbo', }); } main(); STEEL\\_BROWSER skyscanner.net 20:05 AMS 13:45 MAD 20:05 AMS 20:05 AMS 20:05 AMS 20:05 AMS 800B+ Tokens Scraped 800,000+ Browser Hours Served \u003C1s Avg. Session Start Time Use Cases What Developers Build on Steel From large-scale scrape jobs to fully autonomous web agents, Steel makes it easy to run browser automations in the cloud. AI Agents Foundational Model Training AI Shopping Assistants Web Scraping RPA Software Sales Automation QA Software Customer Service Agents Sessions API Spin up on-demand browser sessions with a simple API call. [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) I’m not a robot Auto CAPTCHA solving Built-in CAPTCHA solving that keeps your automation flowing. Proxy and Browser Fingerprinting Simple controls to never worry about getting flagged as a bot again. 500 ms Steel 5000 ms Competition Quick-start Average session starts in less than 1s when client is in same region. Live - 23h:59m Up to 24h Long Sessions Run for a minute or several hours, each session can run up to 24 hours. cookie#1 cookie#4 cookie#3 cookie#2 Manage and Reuse Context Save and inject cookies and local storage to pick up where you left off. \\-puppeteer.launch ({...}) +puppeteer.connect ({...}) 1-line Change to Run Puppeteer Easily run your Puppeteer, Playwright, or Selenium in the cloud. Playing World-class Observability Session Viewer lets you view and debug live or recorded sessions. Sign in Securely Auto Sign-In Securely give your agents access to auth-walled websites and apps. Get Started with Your Favorite Stack Easily integrate with the tools you know and love. We're always working on adding more to this list. [ Docs ](https:\u002F\u002Fdocs.steel.dev\u002F) Python Node.js Puppeteer Playwright Selenium Coming Soon from steel import Steel client = Steel() \\# Create a session session = client.sessions.create() print(f\"Session created: {session.id}\") \\# Your session is now ready to use! \\# Keep this ID handy - you'll need it to connect with automation tools [ view Python SDK ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-python) Cookbook Examples Try out Steel with the brand new Cookbook examples [ View All Examples ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook) Browser Use Learn how to build an AI Agent with Steel and browser-use [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-browser-use-starter) Magnitude Learn how to build an AI Agent with Steel and Magnitude [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-magnitude-starter) Notte Learn how to build an AI Agent with Steel and Notte [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-notte-starter) OpenAI's Computer Use Learn how to create a browser automation agent with Steel and OpenAI's Computer Use [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-oai-computer-use-node-starter) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-oai-computer-use-python-starter) Playwright Learn how to build a starter project for using Playwright with Steel [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-playwright-starter-js) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-playwright-python-starter) Claude's Computer Use Learn how to create a browser automation agent with Steel and Claude's Computer use [ View (TS) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-claude-computer-use-node-starter) [ View (Python) ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-cookbook\u002Ftree\u002Fmain\u002Fexamples\u002Fsteel-claude-computer-use-python-starter) Our Pricing For info, limits and rates, explore a detailed pricing breakdown ↗. Monthly Annually (Save 20%) Hobby Free $10 in Credits per month Everything you need to start automating browsers. Monthly: 100 Browser Hours [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002F) Starter $29 $29 in Credits per month Everything you need to start automating browsers. Monthly: 290 Browser Hours 2.9 GB Proxy Bandwidth 7.2k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Developers Popular $99 $99 in Credits per month Everything you need to launch your project. Monthly: 1,238 Browser Hours 12 GB Proxy Bandwidth 28k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Startups $499 $499 in Credits per month Everything you need to scale your masterpiece. Monthly: 9,980 Browser Hours 166 GB Proxy Bandwidth 166k Captcha Solves [ Get Started ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) Enterprise Spin up thousands of browser sessions on the cloud. [ Book a Demo ](https:\u002F\u002Fcal.com\u002Fhussien-hussien-fjxt3x\u002Fsteel-enterprise-demo) Custom Pricing Higher Limits Discounted Rates Dedicated Clusters & APIs Unlimited Session Concurrency GitHub Repository Steel is proudly open source. Run our sessions locally or self-hosted with our docker container. [ steel-browser ](https:\u002F\u002Fgithub.com\u002Fsteel-dev\u002Fsteel-browser) Research Grants If you’re doing research for a school, send us an email from your school emails for grants in the form of free credits. Email us at research@steel.dev Ready to Build with Steel? [ Start For Free ](https:\u002F\u002Fapp.steel.dev\u002Fsign-up) [ Documentation ](https:\u002F\u002Fdocs.steel.dev\u002F) A better way to take your LLMs online. © Nen Labs · Inc. 2026. Terms of Service · Privacy Policy [ All Systems Operational ](https:\u002F\u002Fstatus.steel.dev\u002F) Platform Docs Status Brand Assets Launch Week @ Steel Join the Community X \u002F Twitter GitHub Blog Discord Careers ![](https:\u002F\u002Fanalytics.twitter.com\u002F1\u002Fi\u002Fadsct?bci=4&dv=UTC%26en-US%26Google%20Inc.%26Linux%20x86_64%26255%26800%26600%264%2624%26800%26600%260%26na&eci=3&event=%7B%7D&event_id=be57c6d2-542b-41e7-8915-59accfbb7df3&integration=advert …. A Clean \u002F Minimalist, Dark Mode, Futuristic \u002F Sci-Fi website in the SaaS & Software industry. The overall color palette features Black, White, Teal, Gray, Yellow. The typography features Geist (Sans Serif), Geist Mono, Geist (Sans Serif), Geist (Sans Serif), Inter (Sans Serif), JetBrains Mono (Sans Serif). Built using Framer Sites, React. AI description: Steel provides a high-performance, open-source headless browser API specifically engineered to power the next generation of AI agents. By allowing developers to control vast fleets of browsers in the cloud, the platform bridges the gap between LLM reasoning and real-world web interaction, enabling seamless automation and data scraping at scale. The visual identity is unapologetically developer-centric, utilizing a sophisticated dark mode aesthetic paired with high-contrast accents. The design employs a gritty, textured background that evokes a sense of raw, industrial power, while clean typography and structured UI components maintain a sense of technical precision. It is a brand built for engineers who value both speed and robust infrastructure."]