{"slug": "google-antigravity", "name": "Google Antigravity", "category": "ai-coding", "category_name": "AI coding tools", "description": "Google's agentic coding tool for Gemini models, offered as a desktop IDE and the agy CLI.", "recommenders": 74, "commenters": 256, "warners": 75, "score": 24.089, "rising": 0.67, "first": "2025-11-18", "latest": "2026-10-30", "why": "Used as a main coding agent that some rank alongside Claude Code, praised for fast Gemini Flash models and for planning and finishing tasks with little supervision.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 78, "neutral": 110, "compare": 12, "warn": 81}, "spark": [6, 7, 10, 11, 10, 2, 4, 7, 2, 5, 11, 3], "listed": true, "loved": ["Main coding agent", "Gets tasks done"], "caveat": "Account hurdles", "short_quote": "After getting frustrated with how slow Claude was on my personal projects, I switched to Google Antigravity and the speed difference is huge.", "kind": "", "icon": true, "domain": "antigravity.google", "github_repo": null, "official_url": "https://antigravity.google/", "pricing": "freemium", "aliases": ["Antigravity", "Google Antigravity", "AntiGravity", "Antigravity IDE"], "verdict": [{"text": "Used as a main coding agent that some rank alongside Claude Code,", "cites": [{"id": 795552320, "item_id": 49722020, "url": "https://news.ycombinator.com/item?id=49722020", "by": "qingcharles", "date": "2026-09-16", "version": null, "stance": "recommend", "quote": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me.", "excerpt": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me. (and I use Claude, Codex, Grok and Muse for all the other projects)", "context": "", "use_case": "main coding agent", "switched_from": "", "story": "Gemini 3.8 Live and 3.8 Live Extended Thinking"}, {"id": 792650080, "item_id": 49540630, "url": "https://news.ycombinator.com/item?id=49540630", "by": "drusepth", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "Antigravity is probably the best of the bunch I've tried.", "excerpt": "Antigravity is probably the best of the bunch I've tried. I'd say it's pretty comparable to Claude Code (I use both daily).", "context": "", "use_case": "harness for Gemini models", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792794016, "item_id": 49549626, "url": "https://news.ycombinator.com/item?id=49549626", "by": "handzhiev", "date": "2026-09-03", "version": null, "stance": "recommend", "quote": "AGY is best probably, yes. That's what I use.", "excerpt": "AGY is best probably, yes. That's what I use. It works with others too, like OpenCode etc.", "context": "", "use_case": "agent harness for Gemini", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}]}, {"text": " praised for fast Gemini Flash models and for planning and finishing tasks with little supervision.", "cites": [{"id": 794771825, "item_id": 49673239, "url": "https://news.ycombinator.com/item?id=49673239", "by": "vallerie", "date": "2026-09-12", "version": null, "stance": "recommend", "quote": "I've found good success with the new Gemini models on Antigravity.", "excerpt": "I've found good success with the new Gemini models on Antigravity. Granted I use my models either: - like a fancy auto complete (here are some stub methods, they should do X, fill them in) - using fairly detailed plans and test harnesses, so blowing up the world is hard The 3.X Flash family have been fairly capable models, and the selling point for me is just raw speed. Gemini is noticeably faster than the competitio", "context": "", "use_case": "coding", "switched_from": "", "story": "Ask HN: What default model do you use and why?"}, {"id": 792623552, "item_id": 49538972, "url": "https://news.ycombinator.com/item?id=49538972", "by": "porridgeraisin", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "agy is good for those cases where you are willing to put the effort into the harness", "excerpt": "agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want a \"general coding assistant\" like codex or claudecode. The reality is that if you optimise a harness for a family of tasks[1], then most of these models give successful output. An", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 795882150, "item_id": 49742634, "url": "https://news.ycombinator.com/item?id=49742634", "by": "vegadw", "date": "2026-09-17", "version": null, "stance": "recommend", "quote": "On Antigravity/Gemini, I can give it a big list-o-TODOs and have it make a very detailed implementation plan itself, carry it out, and generally do the thing, all unsupervised.", "excerpt": "Not really? I have some custom skills/MCP servers I use: one MCP server for using https://usefeyn.com/blog/pulpie-pareto-optimal-models-for-cleaning-the-web/ ran locally so that reading web pages is cheaper and uses up less context One MCP server for doing automated code quality checking, using Valknut, Jscpd, and Lizard to output scores, combine/aggregate them. That's then give to the model so that it can see if the", "context": "", "use_case": "AI coding agent", "switched_from": "", "story": "Show HN: Share your AI Setup, Learn from others"}, {"id": 792294016, "item_id": 49518376, "url": "https://news.ycombinator.com/item?id=49518376", "by": "sanarothe", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "The codebot produces everything I ask", "excerpt": "Also curious to see what issues people are having. I'm a not-software engineer but I use antigravity quite a bit for my own projects. If you give it a task, it completes it. It might so a single agent chug, or delegate some parallel tasks.. The architecture is essentially what was agreed on during the planning and spec phases. So I'm just not understanding like... The codebot produces everything I ask, I use Gemini a", "context": "not a software engineer", "use_case": "personal projects", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}]}, {"text": " The most common complaints are Google account and subscription hurdles", "cites": [{"id": 804591265, "item_id": 50286954, "url": "https://news.ycombinator.com/item?id=50286954", "by": "andy_ppp", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it wanted me to sign into Google to even edit a file. Forget it.", "excerpt": "I downloaded Anti Gravity to give it a try and it wanted me to sign into Google to even edit a file. Forget it.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804496640, "item_id": 50281040, "url": "https://news.ycombinator.com/item?id=50281040", "by": "kisamoto", "date": "2026-10-28", "version": null, "stance": "warn", "quote": "But that leads me to an error page, because Ultra is only available for personal accounts apparently", "excerpt": "I'm on a Workspace Pro subscription and the models I have are (in order of the dropdown on gemini.google.com): - 3.6 Flash - 3.8 Thinking - 3.1 Pro I would love to sign up to Antigravity to try out these new Gemini 4 models but the Antigravity pricing page - https://antigravity.google/pricing - recommends me (a \"Developer\") to \"Get Ultra\". But that leads me to an error page, because Ultra is only available for person", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 792808688, "item_id": 49550543, "url": "https://news.ycombinator.com/item?id=49550543", "by": "deno", "date": "2026-09-03", "version": null, "stance": "warn", "quote": "still you won't be able to use even the free tier of Antigravity.", "excerpt": "Also word of warning: don't buy Google AI subscription before you confirm you have \"Antigravity entitlement\" whatever that is. You can have \"verified\" account (for 18+ and Gmail), bank card added, active AI Pro subscription, and still you won't be able to use even the free tier of Antigravity. And their verification is some major AI slop that just doesn't work. This is open from June but this was already a bug in Gem", "context": "", "use_case": "", "switched_from": "", "story": "Google Antigravity TOS: 3rd party usage can get Google account suspended"}]}, {"text": " and usage quotas that run out quickly.", "cites": [{"id": 804546704, "item_id": 50284169, "url": "https://news.ycombinator.com/item?id=50284169", "by": "hackernudes", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it used up my weekly quota in 10 minutes without stopping or giving me anything useful at all", "excerpt": "Me too. I have a basic Google One account which is now called \"Google AI Plus\" and it says \"Get 2x higher usage access than Free\". The Google AI Pro plan says \"Get 4x higher usage access than Free\" (from https://gemini.google/subscriptions/). I tried antigravity with Gemini 3.8 Flash High. I used a prompt similar to what I use with Codex and Claude Code and it used up my weekly quota in 10 minutes without stopping or", "context": "embedded firmware developer, pays $20 plans for Anthropic and OpenAI", "use_case": "embedded firmware coding", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 790646352, "item_id": 49415397, "url": "https://news.ycombinator.com/item?id=49415397", "by": "DefamationStati", "date": "2026-08-24", "version": null, "stance": "warn", "quote": "the limits as absolutely abysmal when using 3.7 flash in antigravity", "excerpt": "I've got the AI pro plan for Gemini and the limits as absolutely abysmal when using 3.7 flash in antigravity, If I use swarms then I can't even finish a single prompt without it reaching the limit.", "context": "has Gemini AI Pro plan", "use_case": "AI coding", "switched_from": "", "story": "Anthropic's best AI model struggles to attract users as cheaper tools thrive"}]}], "praise": [{"label": "Main coding agent", "summary": "Several commenters use it as their primary agent and call it among the best they've tried, comparable to Claude Code.", "quotes": [{"id": 795552320, "item_id": 49722020, "url": "https://news.ycombinator.com/item?id=49722020", "by": "qingcharles", "date": "2026-09-16", "version": null, "stance": "recommend", "quote": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me.", "excerpt": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me. (and I use Claude, Codex, Grok and Muse for all the other projects)", "context": "", "use_case": "main coding agent", "switched_from": "", "story": "Gemini 3.8 Live and 3.8 Live Extended Thinking"}, {"id": 792650080, "item_id": 49540630, "url": "https://news.ycombinator.com/item?id=49540630", "by": "drusepth", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "Antigravity is probably the best of the bunch I've tried.", "excerpt": "Antigravity is probably the best of the bunch I've tried. I'd say it's pretty comparable to Claude Code (I use both daily).", "context": "", "use_case": "harness for Gemini models", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792794016, "item_id": 49549626, "url": "https://news.ycombinator.com/item?id=49549626", "by": "handzhiev", "date": "2026-09-03", "version": null, "stance": "recommend", "quote": "AGY is best probably, yes. That's what I use.", "excerpt": "AGY is best probably, yes. That's what I use. It works with others too, like OpenCode etc.", "context": "", "use_case": "agent harness for Gemini", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}]}, {"label": "Gets tasks done", "summary": "Users say it plans and carries out work, from TODO lists to personal projects, often without supervision.", "quotes": [{"id": 795882150, "item_id": 49742634, "url": "https://news.ycombinator.com/item?id=49742634", "by": "vegadw", "date": "2026-09-17", "version": null, "stance": "recommend", "quote": "On Antigravity/Gemini, I can give it a big list-o-TODOs and have it make a very detailed implementation plan itself, carry it out, and generally do the thing, all unsupervised.", "excerpt": "Not really? I have some custom skills/MCP servers I use: one MCP server for using https://usefeyn.com/blog/pulpie-pareto-optimal-models-for-cleaning-the-web/ ran locally so that reading web pages is cheaper and uses up less context One MCP server for doing automated code quality checking, using Valknut, Jscpd, and Lizard to output scores, combine/aggregate them. That's then give to the model so that it can see if the", "context": "", "use_case": "AI coding agent", "switched_from": "", "story": "Show HN: Share your AI Setup, Learn from others"}, {"id": 792294016, "item_id": 49518376, "url": "https://news.ycombinator.com/item?id=49518376", "by": "sanarothe", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "The codebot produces everything I ask", "excerpt": "Also curious to see what issues people are having. I'm a not-software engineer but I use antigravity quite a bit for my own projects. If you give it a task, it completes it. It might so a single agent chug, or delegate some parallel tasks.. The architecture is essentially what was agreed on during the planning and spec phases. So I'm just not understanding like... The codebot produces everything I ask, I use Gemini a", "context": "not a software engineer", "use_case": "personal projects", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}, {"id": 792294848, "item_id": 49518428, "url": "https://news.ycombinator.com/item?id=49518428", "by": "doix", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "I just stick with antigravity because it works for me", "excerpt": "I use it for all my non-professional work and I find it works well enough for me. Mostly for things that I know are possible and have a vague idea of how I would do it, but it would take me longer. Some examples in no particular order: * Made my wife's travel blog/site * Chrome extension for Crunchyroll to allow multisubs + local translation for unsupported languages (so my wife and I can watch and she doesn't need t", "context": "non-professional projects", "use_case": "personal coding projects", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}]}, {"label": "Speed", "summary": "Gemini Flash models in Antigravity are described as noticeably faster than the competition.", "quotes": [{"id": 794771825, "item_id": 49673239, "url": "https://news.ycombinator.com/item?id=49673239", "by": "vallerie", "date": "2026-09-12", "version": null, "stance": "recommend", "quote": "I've found good success with the new Gemini models on Antigravity.", "excerpt": "I've found good success with the new Gemini models on Antigravity. Granted I use my models either: - like a fancy auto complete (here are some stub methods, they should do X, fill them in) - using fairly detailed plans and test harnesses, so blowing up the world is hard The 3.X Flash family have been fairly capable models, and the selling point for me is just raw speed. Gemini is noticeably faster than the competitio", "context": "", "use_case": "coding", "switched_from": "", "story": "Ask HN: What default model do you use and why?"}, {"id": 792623552, "item_id": 49538972, "url": "https://news.ycombinator.com/item?id=49538972", "by": "porridgeraisin", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "agy is good for those cases where you are willing to put the effort into the harness", "excerpt": "agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want a \"general coding assistant\" like codex or claudecode. The reality is that if you optimise a harness for a family of tasks[1], then most of these models give successful output. An", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}]}, {"label": "Model choice", "summary": "Commenters pair Gemini 3.8 Flash with Opus inside Antigravity and say Gemini holds its own.", "quotes": [{"id": 798621665, "item_id": 49913854, "url": "https://news.ycombinator.com/item?id=49913854", "by": "spankalee", "date": "2026-09-30", "version": null, "stance": "recommend", "quote": "so is the Antigravity harness", "excerpt": "3.8 Flash is just quite good, and so is the Antigravity harness. I use a mix of Fable 5.1, Opus 5.5, and Gemini 3.8 Flash and Gemini holds it's own. Especially in writing, frontend, and sysadmin work. agy for configuring a NixOS system has been truly incredible.", "context": "", "use_case": "coding harness", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 794781905, "item_id": 49673869, "url": "https://news.ycombinator.com/item?id=49673869", "by": "sidibe", "date": "2026-09-12", "version": null, "stance": "recommend", "quote": "Gemini 3.8 + antigravity. Opus also in antigravity for harder stuff.", "excerpt": "Gemini 3.8 + antigravity. Opus also in antigravity for harder stuff.", "context": "", "use_case": "coding harness", "switched_from": "", "story": "Ask HN: What default model do you use and why?"}]}, {"label": "Parallel sessions", "summary": "One user runs 5-6 concurrent sessions and finds it largely fine.", "quotes": [{"id": 804487888, "item_id": 50280493, "url": "https://news.ycombinator.com/item?id=50280493", "by": "nostrebored", "date": "2026-10-28", "version": null, "stance": "recommend", "quote": "I\u2019ve been using maybe 5-6 concurrent sessions and it\u2019s largely fine", "excerpt": "yes, I\u2019ve only had it really degrade once and then I used a different agy session to fix the session table. I\u2019ve been using maybe 5-6 concurrent sessions and it\u2019s largely fine", "context": "", "use_case": "CLI coding agent sessions", "switched_from": "", "story": "Gemini 4 Carbon"}]}, {"label": "Custom harnesses", "summary": "It shines when you invest in evals, hooks and custom tools for a specific family of tasks.", "quotes": [{"id": 792623552, "item_id": 49538972, "url": "https://news.ycombinator.com/item?id=49538972", "by": "porridgeraisin", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "agy is good for those cases where you are willing to put the effort into the harness", "excerpt": "agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want a \"general coding assistant\" like codex or claudecode. The reality is that if you optimise a harness for a family of tasks[1], then most of these models give successful output. An", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}]}], "caveats": [{"label": "Account hurdles", "summary": "Commenters hit required Google sign-in, Workspace accounts blocked from Ultra, and entitlement bugs even on paid plans.", "quotes": [{"id": 804591265, "item_id": 50286954, "url": "https://news.ycombinator.com/item?id=50286954", "by": "andy_ppp", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it wanted me to sign into Google to even edit a file. Forget it.", "excerpt": "I downloaded Anti Gravity to give it a try and it wanted me to sign into Google to even edit a file. Forget it.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804496640, "item_id": 50281040, "url": "https://news.ycombinator.com/item?id=50281040", "by": "kisamoto", "date": "2026-10-28", "version": null, "stance": "warn", "quote": "But that leads me to an error page, because Ultra is only available for personal accounts apparently", "excerpt": "I'm on a Workspace Pro subscription and the models I have are (in order of the dropdown on gemini.google.com): - 3.6 Flash - 3.8 Thinking - 3.1 Pro I would love to sign up to Antigravity to try out these new Gemini 4 models but the Antigravity pricing page - https://antigravity.google/pricing - recommends me (a \"Developer\") to \"Get Ultra\". But that leads me to an error page, because Ultra is only available for person", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 792808688, "item_id": 49550543, "url": "https://news.ycombinator.com/item?id=49550543", "by": "deno", "date": "2026-09-03", "version": null, "stance": "warn", "quote": "still you won't be able to use even the free tier of Antigravity.", "excerpt": "Also word of warning: don't buy Google AI subscription before you confirm you have \"Antigravity entitlement\" whatever that is. You can have \"verified\" account (for 18+ and Gmail), bank card added, active AI Pro subscription, and still you won't be able to use even the free tier of Antigravity. And their verification is some major AI slop that just doesn't work. This is open from June but this was already a bug in Gem", "context": "", "use_case": "", "switched_from": "", "story": "Google Antigravity TOS: 3rd party usage can get Google account suspended"}]}, {"label": "Usage limits", "summary": "Paid Google AI plans' quotas reportedly ran out within minutes or a single prompt.", "quotes": [{"id": 804546704, "item_id": 50284169, "url": "https://news.ycombinator.com/item?id=50284169", "by": "hackernudes", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it used up my weekly quota in 10 minutes without stopping or giving me anything useful at all", "excerpt": "Me too. I have a basic Google One account which is now called \"Google AI Plus\" and it says \"Get 2x higher usage access than Free\". The Google AI Pro plan says \"Get 4x higher usage access than Free\" (from https://gemini.google/subscriptions/). I tried antigravity with Gemini 3.8 Flash High. I used a prompt similar to what I use with Codex and Claude Code and it used up my weekly quota in 10 minutes without stopping or", "context": "embedded firmware developer, pays $20 plans for Anthropic and OpenAI", "use_case": "embedded firmware coding", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 790646352, "item_id": 49415397, "url": "https://news.ycombinator.com/item?id=49415397", "by": "DefamationStati", "date": "2026-08-24", "version": null, "stance": "warn", "quote": "the limits as absolutely abysmal when using 3.7 flash in antigravity", "excerpt": "I've got the AI pro plan for Gemini and the limits as absolutely abysmal when using 3.7 flash in antigravity, If I use swarms then I can't even finish a single prompt without it reaching the limit.", "context": "has Gemini AI Pro plan", "use_case": "AI coding", "switched_from": "", "story": "Anthropic's best AI model struggles to attract users as cheaper tools thrive"}]}, {"label": "Desktop UX", "summary": "Critics call the desktop app unusable or heavy, and say the experience got worse after launch.", "quotes": [{"id": 792294288, "item_id": 49518393, "url": "https://news.ycombinator.com/item?id=49518393", "by": "dan_voronov", "date": "2026-09-01", "version": null, "stance": "warn", "quote": "The desktop version was basically unusable, so I used the console one.", "excerpt": "I suffered with Antigravity 1 and 2 for a long time (I\u2019m not some weirdo, I just run a blog where I do deep dives into AI tools for programming=). The desktop version was basically unusable, so I used the console one. A couple of weeks ago I finally dropped it and switched to Pi with antigr login. With plugins you can configure it however you want. I just installed one of a deep-search plugins too: you need to add AP", "context": "runs a blog doing deep dives into AI programming tools", "use_case": "AI coding", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}, {"id": 801584178, "item_id": 50099011, "url": "https://news.ycombinator.com/item?id=50099011", "by": "usernametaken29", "date": "2026-10-15", "version": null, "stance": "warn", "quote": "it\u2019s about as heavy as antigravity", "excerpt": "I wish they\u2019d invest more time in performance improvements over code mods. I recently switched to pi which uses an order of magnitude less RAM (hundreds vs dozens) and noticeably less CPU on my machine\u2026 OpenCode is really nice but it\u2019s about as heavy as antigravity and is substandard in features\u2026", "context": "", "use_case": "", "switched_from": "", "story": "OpenCode 2.0"}, {"id": 792287120, "item_id": 49517945, "url": "https://news.ycombinator.com/item?id=49517945", "by": "CSMastermind", "date": "2026-09-01", "version": null, "stance": "warn", "quote": "There was a brief period of time I was using Anti-gravity as my daily driver when they first introduced it and then they just completely screwed it up.", "excerpt": "It's impressive how badly Google has screwed up their models and coding products. There was a brief period of time I was using Anti-gravity as my daily driver when they first introduced it and then they just completely screwed it up.", "context": "", "use_case": "daily coding driver", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}]}, {"label": "Permission prompts", "summary": "Without Turbo mode, frequent permission prompts make getting work done hard.", "quotes": [{"id": 798723008, "item_id": 49920188, "url": "https://news.ycombinator.com/item?id=49920188", "by": "smartbit", "date": "2026-10-01", "version": null, "stance": "warn", "quote": "still many prompts making getting-work-done impossible", "excerpt": "do you mean? accept-edits also available with shift-tab in agy cli [0]. That is not related to executing commands, only to allowing agy edit files. Unless you set Turbo-mode == yolo-mode, agy gui prompts a zillion times too. With 'some features added' I'm referencing this but don't see change in daily work (still many prompts making getting-work-done impossible): v2.14.0 (September 15, 2026) \"New Permissions System\" ", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}]}, {"label": "Prompt injection", "summary": "A published data-exfiltration attack via prompt injection is cited as a security risk.", "quotes": [{"id": 804728080, "item_id": 50295505, "url": "https://news.ycombinator.com/item?id=50295505", "by": "ShinyLeftPad", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "Here's prompt injection via a harness for example", "excerpt": "Here's prompt injection via a harness for example: https://www.promptarmor.com/resources/google-antigravity-exfiltrates-data took me about 30 seconds to find. there's tons of them. of course it's an arms race, the models get better the harnesses get better the attackers get better but again: for the model itself, your prompt and third party data it reads are the same thing (input). as long as it is true, the attack w", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Auto Dismiss Mac OS \"Allow <app> to find devices on local networks?\""}]}, {"label": "Not general-purpose", "summary": "Some find it weak as a general coding assistant compared with Codex or Claude Code.", "quotes": [{"id": 792623552, "item_id": 49538972, "url": "https://news.ycombinator.com/item?id=49538972", "by": "porridgeraisin", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "agy is good for those cases where you are willing to put the effort into the harness", "excerpt": "agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want a \"general coding assistant\" like codex or claudecode. The reality is that if you optimise a harness for a family of tasks[1], then most of these models give successful output. An", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 804106355, "item_id": 50256647, "url": "https://news.ycombinator.com/item?id=50256647", "by": "dizhn", "date": "2026-10-27", "version": null, "stance": "warn", "quote": "Both antigravity and agy are toys compared to what has been available for at least two years.", "excerpt": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me. And I'm not ready to keep throwing money at it for it to become usable. Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing and a very generous side items package. I use the models but not the \"harness\" though. Not directly anyway. I am still handicapped b", "context": "", "use_case": "", "switched_from": "", "story": "Anthropic S-1"}]}], "who": [{"by": "vardhan1237", "context": "PaperPilot developer", "url": "https://news.ycombinator.com/item?id=50242458", "date": "2026-10-26"}, {"by": "harshalizee", "context": "works in an MS-stack company as a dotnet developer", "url": "https://news.ycombinator.com/item?id=50133349", "date": "2026-10-17"}, {"by": "mceleri", "context": "developer", "url": "https://news.ycombinator.com/item?id=50012420", "date": "2026-10-08"}, {"by": "rakyll", "context": "co-creator of AX", "url": "https://news.ycombinator.com/item?id=49793311", "date": "2026-09-21"}, {"by": "doix", "context": "non-professional projects", "url": "https://news.ycombinator.com/item?id=49518428", "date": "2026-09-01"}, {"by": "sanarothe", "context": "not a software engineer", "url": "https://news.ycombinator.com/item?id=49518376", "date": "2026-09-01"}, {"by": "hn_throw2025", "context": "has a Gemini subscription", "url": "https://news.ycombinator.com/item?id=49487793", "date": "2026-08-29"}, {"by": "antonvs", "context": "works at a company that values cheap tooling", "url": "https://news.ycombinator.com/item?id=49191709", "date": "2026-08-05"}, {"by": "algoth1", "context": "non-coding professional automating complex tasks", "url": "https://news.ycombinator.com/item?id=48693357", "date": "2026-06-26"}, {"by": "trollbridge", "context": "small dev team of 6", "url": "https://news.ycombinator.com/item?id=48648590", "date": "2026-06-23"}, {"by": "shsh1312", "context": "low-budget AI coding", "url": "https://news.ycombinator.com/item?id=48594703", "date": "2026-06-18"}, {"by": "matbanik", "context": "Linux user, QA work", "url": "https://news.ycombinator.com/item?id=48536053", "date": "2026-06-14"}], "tags": ["for personal projects", "for non-engineers", "for NixOS config", "for task-specific harnesses"], "use_cases": ["ai coding", "coding", "ai coding ide", "ai coding agent", "coding agent", "ai ide"], "compared": [{"slug": "claude-code", "name": "Claude Code", "co": 56, "switched_from": 0, "switched_to": 1}, {"slug": "codex", "name": "Codex", "co": 48, "switched_from": 0, "switched_to": 3}, {"slug": "gemini", "name": "Gemini", "co": 46, "switched_from": 0, "switched_to": 0}, {"slug": "cursor", "name": "Cursor", "co": 31, "switched_from": 3, "switched_to": 1}, {"slug": "gemini-cli", "name": "Gemini CLI", "co": 27, "switched_from": 2, "switched_to": 0}, {"slug": "claude", "name": "Claude", "co": 27, "switched_from": 1, "switched_to": 0}, {"slug": "claude-opus", "name": "Claude Opus", "co": 22, "switched_from": 0, "switched_to": 0}, {"slug": "gemini-pro", "name": "Gemini Pro", "co": 17, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=45967814", "title": "Google Antigravity", "mentions": 22}, {"url": "https://news.ycombinator.com/item?id=48222529", "title": "Google's Antigravity Bait and Switch", "mentions": 13}, {"url": "https://news.ycombinator.com/item?id=49913571", "title": "Gemini 4 Argon", "mentions": 8}, {"url": "https://news.ycombinator.com/item?id=47195371", "title": "Addressing Antigravity Bans and Reinstating Access", "mentions": 6}, {"url": "https://news.ycombinator.com/item?id=50276792", "title": "Gemini 4 Carbon", "mentions": 5}, {"url": "https://news.ycombinator.com/item?id=46048996", "title": "Google Antigravity exfiltrates data via indirect prompt injection attack", "mentions": 5}], "sentiment_by_month": {"2025-11": {"neutral": 12, "warn": 18, "recommend": 6}, "2025-12": {"neutral": 4, "warn": 7, "recommend": 7}, "2026-01": {"neutral": 13, "recommend": 10, "warn": 3}, "2026-02": {"neutral": 16, "warn": 5, "recommend": 11, "compare": 3}, "2026-03": {"neutral": 5, "recommend": 10, "warn": 5, "compare": 3}, "2026-04": {"neutral": 9, "recommend": 2, "warn": 1}, "2026-05": {"recommend": 4, "neutral": 13, "warn": 18, "compare": 3}, "2026-06": {"neutral": 7, "recommend": 7, "warn": 4}, "2026-07": {"recommend": 2, "warn": 3, "neutral": 1}, "2026-08": {"neutral": 4, "recommend": 5, "warn": 3}, "2026-09": {"warn": 4, "recommend": 11, "neutral": 12}, "2026-10": {"warn": 10, "neutral": 14, "compare": 3, "recommend": 3}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 804849441, "item_id": 50303090, "url": "https://news.ycombinator.com/item?id=50303090", "by": "darkwi11ow", "date": "2026-10-30", "version": null, "stance": "compare", "quote": "Google tries to compete via Antigravity, but that also uses OpenVSX.", "excerpt": "There is no alternative really. Cursor was the market leader in LLM-assisted IDE experience before SpaceX acquisition, though Elon nerfed the product a lot in recent months. And there is still no good alternative to it. Google tries to compete via Antigravity, but that also uses OpenVSX. I myself use Cursor and use Claude there installed via OpenVSX, and was not affected only thanks to my laziness (I saw upgrade butt", "context": "", "use_case": "AI-assisted IDE", "switched_from": "", "story": "Backdoor in Claude Code for VS Code extension on the Cursor Marketplace"}, {"id": 804728080, "item_id": 50295505, "url": "https://news.ycombinator.com/item?id=50295505", "by": "ShinyLeftPad", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "Here's prompt injection via a harness for example", "excerpt": "Here's prompt injection via a harness for example: https://www.promptarmor.com/resources/google-antigravity-exfiltrates-data took me about 30 seconds to find. there's tons of them. of course it's an arms race, the models get better the harnesses get better the attackers get better but again: for the model itself, your prompt and third party data it reads are the same thing (input). as long as it is true, the attack w", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Auto Dismiss Mac OS \"Allow <app> to find devices on local networks?\""}, {"id": 804591265, "item_id": 50286954, "url": "https://news.ycombinator.com/item?id=50286954", "by": "andy_ppp", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it wanted me to sign into Google to even edit a file. Forget it.", "excerpt": "I downloaded Anti Gravity to give it a try and it wanted me to sign into Google to even edit a file. Forget it.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804566802, "item_id": 50285425, "url": "https://news.ycombinator.com/item?id=50285425", "by": "spockz", "date": "2026-10-29", "version": null, "stance": "neutral", "quote": "I\u2019m using the default anti gravity setup.", "excerpt": "I\u2019m using the default anti gravity setup. 3.8 medium is my daily driver. This gets me pretty far. (Definitely not as far as Luna and Sol). When I ask the same question to Argon it is 50% of my weekly usage limit gone. So either the usage limit is really a lot smaller or somehow argon thinks a whole lot more even for simple tasks.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804546704, "item_id": 50284169, "url": "https://news.ycombinator.com/item?id=50284169", "by": "hackernudes", "date": "2026-10-29", "version": null, "stance": "warn", "quote": "it used up my weekly quota in 10 minutes without stopping or giving me anything useful at all", "excerpt": "Me too. I have a basic Google One account which is now called \"Google AI Plus\" and it says \"Get 2x higher usage access than Free\". The Google AI Pro plan says \"Get 4x higher usage access than Free\" (from https://gemini.google/subscriptions/). I tried antigravity with Gemini 3.8 Flash High. I used a prompt similar to what I use with Codex and Claude Code and it used up my weekly quota in 10 minutes without stopping or", "context": "embedded firmware developer, pays $20 plans for Anthropic and OpenAI", "use_case": "embedded firmware coding", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804496640, "item_id": 50281040, "url": "https://news.ycombinator.com/item?id=50281040", "by": "kisamoto", "date": "2026-10-28", "version": null, "stance": "warn", "quote": "But that leads me to an error page, because Ultra is only available for personal accounts apparently", "excerpt": "I'm on a Workspace Pro subscription and the models I have are (in order of the dropdown on gemini.google.com): - 3.6 Flash - 3.8 Thinking - 3.1 Pro I would love to sign up to Antigravity to try out these new Gemini 4 models but the Antigravity pricing page - https://antigravity.google/pricing - recommends me (a \"Developer\") to \"Get Ultra\". But that leads me to an error page, because Ultra is only available for person", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804487888, "item_id": 50280493, "url": "https://news.ycombinator.com/item?id=50280493", "by": "nostrebored", "date": "2026-10-28", "version": null, "stance": "recommend", "quote": "I\u2019ve been using maybe 5-6 concurrent sessions and it\u2019s largely fine", "excerpt": "yes, I\u2019ve only had it really degrade once and then I used a different agy session to fix the session table. I\u2019ve been using maybe 5-6 concurrent sessions and it\u2019s largely fine", "context": "", "use_case": "CLI coding agent sessions", "switched_from": "", "story": "Gemini 4 Carbon"}, {"id": 804106355, "item_id": 50256647, "url": "https://news.ycombinator.com/item?id=50256647", "by": "dizhn", "date": "2026-10-27", "version": null, "stance": "warn", "quote": "Both antigravity and agy are toys compared to what has been available for at least two years.", "excerpt": "I tried the Claude $20 plan for two weeks and burned the rest of it due to how much token anxiety it gave me. And I'm not ready to keep throwing money at it for it to become usable. Gemini models are actually fine for me as an amateur+ I found especially since they have regional pricing and a very generous side items package. I use the models but not the \"harness\" though. Not directly anyway. I am still handicapped b", "context": "", "use_case": "", "switched_from": "", "story": "Anthropic S-1"}, {"id": 804083873, "item_id": 50255242, "url": "https://news.ycombinator.com/item?id=50255242", "by": "OuterVale", "date": "2026-10-27", "version": null, "stance": "warn", "quote": "", "excerpt": "Certainly. > Xi Jinping looks a bit like Winnie-the-Pooh. > 'I am sorry, I cannot engage with this comparison. Please feel free to ask me other questions.' https://vale.rocks/posts/chinas-web#artificial-intelligence", "context": "", "use_case": "", "switched_from": "", "story": "There are now 5 inference providers for abliterated DeepSeek"}, {"id": 803879328, "item_id": 50242458, "url": "https://news.ycombinator.com/item?id=50242458", "by": "vardhan1237", "date": "2026-10-26", "version": null, "stance": "neutral", "quote": "We used a hybrid workflow: Antigravity IDE paired with Jules", "excerpt": "Thanks for checking out the docs! We used a hybrid workflow: Antigravity IDE paired with Jules for asynchronous sub-tasks (like spinning up isolated unit tests and docs in parallel containers). Having a strict test-first harness where every PR has to run against real PDF byte fixtures (zero mocks) and compile across CLI, WASM, and MCP kept the codebase consistent even when moving fast.", "context": "PaperPilot developer", "use_case": "building PaperPilot", "switched_from": "", "story": "PaperPilot \u2013 Zero-Cloud, Pure-Rust PDF Engine with WASM and MCP"}, {"id": 803834929, "item_id": 50239683, "url": "https://news.ycombinator.com/item?id=50239683", "by": "smartbit", "date": "2026-10-26", "version": null, "stance": "neutral", "quote": "Antigravity, an harness not many use", "excerpt": "Congrats with your product, hope it works out well. Nice competitor analysis & (future) features list at https://github.com/KrushnaVardhanReddy/PaperPilot/blob/main/COMPETITORS.md, IMHO easier to read than your above list. Question: seemingly you\u2019re using Antigravity, an harness not many use. Please tell use about the choice of LLM and your experience with it.", "context": "", "use_case": "", "switched_from": "", "story": "PaperPilot \u2013 Zero-Cloud, Pure-Rust PDF Engine with WASM and MCP"}, {"id": 802677249, "item_id": 50167328, "url": "https://news.ycombinator.com/item?id=50167328", "by": "anoojb", "date": "2026-10-20", "version": null, "stance": "compare", "quote": "The extra baseline quota with Anti-gravity could be worth it.", "excerpt": "I've heard that folks using using a mid-tier Anthropic model like Sonnet 5.5 easily run into limits. When I use GitHub Copilot (for work) with Anthropic models I can easily rack up quite a significant bill. This is why I'm comparing value for the money. While Google may not have the \"best\" models, what's the point if you hit a wall with Claude Code? The extra baseline quota with Anti-gravity could be worth it.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Ask HN: Best value/money AI coding plan bundle?"}, {"id": 802133589, "item_id": 50133349, "url": "https://news.ycombinator.com/item?id=50133349", "by": "harshalizee", "date": "2026-10-17", "version": null, "stance": "recommend", "quote": "primarily use Claude and antigravity", "excerpt": "You would severely limit yourself if you're strict about your list without nuance. GitHub is pretty much bog standard in most companies unless you're working for the few using mercurial, etc. I work in a MS stack and most of your list is untrue for my case. We use slack for messaging, Confluence, Wikimedia, and GDrive for documentation, Google Mail for email, I develop primarily dotnet/C# apps on a MacBook pro M5, ha", "context": "works in an MS-stack company as a dotnet developer", "use_case": "development", "switched_from": "", "story": "Microsoft Teams is a red flag"}, {"id": 801850592, "item_id": 50115662, "url": "https://news.ycombinator.com/item?id=50115662", "by": "fullstackwife", "date": "2026-10-16", "version": null, "stance": "neutral", "quote": "I used 50% in just one task(3d gamedev) on AI Pro in Antigravity desktop.", "excerpt": "I used 50% in just one task(3d gamedev) on AI Pro in Antigravity desktop. It is not my daily driver anyways, but it was nice to see that it is actually performing and completing the task.", "context": "", "use_case": "3D game development", "switched_from": "", "story": "Tell HN: Google Argon on a Pro subscription is worthless"}, {"id": 801813777, "item_id": 50113361, "url": "https://news.ycombinator.com/item?id=50113361", "by": "no-name-here", "date": "2026-10-16", "version": null, "stance": "neutral", "quote": "using Antigravity for some audio software work", "excerpt": "This seems to be his scuba software: https://subsurface-divelog.org It looks like it started 2011. I didn\u2019t check recent commits. I believe Linus also got in tech news in the last year or so for using Antigravity for some audio software work.", "context": "", "use_case": "", "switched_from": "", "story": "Linus Torvalds's first LLM-assisted patch series"}, {"id": 801584178, "item_id": 50099011, "url": "https://news.ycombinator.com/item?id=50099011", "by": "usernametaken29", "date": "2026-10-15", "version": null, "stance": "warn", "quote": "it\u2019s about as heavy as antigravity", "excerpt": "I wish they\u2019d invest more time in performance improvements over code mods. I recently switched to pi which uses an order of magnitude less RAM (hundreds vs dozens) and noticeably less CPU on my machine\u2026 OpenCode is really nice but it\u2019s about as heavy as antigravity and is substandard in features\u2026", "context": "", "use_case": "", "switched_from": "", "story": "OpenCode 2.0"}, {"id": 801416080, "item_id": 50088505, "url": "https://news.ycombinator.com/item?id=50088505", "by": "DuncanCoffee", "date": "2026-10-14", "version": null, "stance": "neutral", "quote": "But is it the Antigravity app, cli, plugin, IDE or SDK?", "excerpt": "But is it the Antigravity app, cli, plugin, IDE or SDK? https://antigravity.google/download", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 801299200, "item_id": 50081200, "url": "https://news.ycombinator.com/item?id=50081200", "by": "astlouis44", "date": "2026-10-14", "version": null, "stance": "neutral", "quote": "Just gemini as antigravity requires Ultra subscription to use it in that IDE", "excerpt": "Just gemini as antigravity requires Ultra subscription to use it in that IDE, whereas Argon is/was available for me on web, so I used the latter", "context": "", "use_case": "Gemini-based IDE", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 801214832, "item_id": 50075927, "url": "https://news.ycombinator.com/item?id=50075927", "by": "I_am_tiberius", "date": "2026-10-13", "version": null, "stance": "neutral", "quote": "Then I found Antigravity, which apparently is Google\u2019s alternative to Codex, and I installed it.", "excerpt": "I just created a Google account, which was a hassle. I tried to find an app and found AI Studio, but I\u2019m not sure what that is. Then I found Antigravity, which apparently is Google\u2019s alternative to Codex, and I installed it. Then I tried to buy the small Gemini AI Ultra package, but it didn\u2019t let me pay. It accepted neither my credit card nor my PayPal account. It\u2019s crazy how bad the Google experience is nowadays. Yo", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 801194032, "item_id": 50074627, "url": "https://news.ycombinator.com/item?id=50074627", "by": "jonesy827", "date": "2026-10-13", "version": null, "stance": "neutral", "quote": "", "excerpt": "Interesting, Google's own article says it's rolling out today to Ultra subscribers. No mention of Pro. https://antigravity.google/blog/gemini-4-argon-in-google-antigravity", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 800792738, "item_id": 50049546, "url": "https://news.ycombinator.com/item?id=50049546", "by": "joaoviniciusgs", "date": "2026-10-12", "version": null, "stance": "neutral", "quote": "I built this using Antigravity and Gemini 3.8 Flash.", "excerpt": "So yeah, I made a clone of Boom 3D. If you don't know what Boom 3D is, it's an audio enhancement app available on Steam and other stores that makes your headphones sound more immersive with 3D spatial audio and other audio effects. It's called OpenBoom, and it's an open-source alternative for Windows built with Rust and Tauri. It aims to bring features like 3D spatial audio, a 31-band equalizer, AutoEq headphone prof", "context": "", "use_case": "building the app", "switched_from": "", "story": "I made an open-source Boom 3D clone. Wanna check it out?"}, {"id": 800749920, "item_id": 50046870, "url": "https://news.ycombinator.com/item?id=50046870", "by": "curiouscat12345", "date": "2026-10-11", "version": null, "stance": "neutral", "quote": "trusted-automation-with-google-antigravity-scaling-secure-finance-integrations-from-40-days-to-5", "excerpt": "https://discuss.google.dev/t/trusted-automation-with-google-antigravity-scaling-secure-finance-integrations-from-40-days-to-5/383313", "context": "", "use_case": "", "switched_from": "", "story": "Terence Tao: Math 2.0 [pdf]"}, {"id": 800596480, "item_id": 50037280, "url": "https://news.ycombinator.com/item?id=50037280", "by": "timonoko", "date": "2026-10-10", "version": null, "stance": "recommend", "quote": "Antigravity made simple Blender-tool that rounds all edges in STL-file.", "excerpt": "For these eternal Fillet-Whiners: Antigravity made simple Blender-tool that rounds all edges in STL-file. Sharp edges never look good in print and avoiding them feels often like too much extra work. https://github.com/timonoko/OpenSCAD-rounding-in-Blender", "context": "", "use_case": "generating Blender tool", "switched_from": "", "story": "OpenSCAD the Programmers Solid 3D CAD Modeller"}, {"id": 800574512, "item_id": 50035907, "url": "https://news.ycombinator.com/item?id=50035907", "by": "simontwiggyahuk", "date": "2026-10-10", "version": null, "stance": "neutral", "quote": "https://discuss.google.dev/t/trusted-automation-with-google-antigravity-scaling-secure-finance-integrations-from-40-days-to-5/383313", "excerpt": "https://discuss.google.dev/t/trusted-automation-with-google-antigravity-scaling-secure-finance-integrations-from-40-days-to-5/383313", "context": "", "use_case": "", "switched_from": "", "story": "Rampart: Browser native on-device PII radaction"}, {"id": 800198721, "item_id": 50012420, "url": "https://news.ycombinator.com/item?id=50012420", "by": "mceleri", "date": "2026-10-08", "version": null, "stance": "neutral", "quote": "Dev guy here, so tailscale + Antigravity (+ GitHub).", "excerpt": "Dev guy here, so tailscale + Antigravity (+ GitHub). I use it to vibecode from my smartphone", "context": "developer", "use_case": "vibecoding from smartphone", "switched_from": "", "story": "Ask HN: What do you run on a $5 VPS that's worth keeping online 24/7?"}, {"id": 799280720, "item_id": 49955045, "url": "https://news.ycombinator.com/item?id=49955045", "by": "dleslie", "date": "2026-10-04", "version": null, "stance": "neutral", "quote": "I don't see a download link for that:", "excerpt": "I don't see a download link for that: https://antigravity.google/docs/ide/extensions/ What I see is integrations for specific editors.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798986672, "item_id": 49936667, "url": "https://news.ycombinator.com/item?id=49936667", "by": "andriy_koval", "date": "2026-10-02", "version": null, "stance": "compare", "quote": "I use agy and codex, and don't see strong difference in ui quality.", "excerpt": "It maybe depend on workflow, you hit some corner cases for agy which I don't hit. I use agy and codex, and don't see strong difference in ui quality.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Pi 1.0"}, {"id": 798861889, "item_id": 49928868, "url": "https://news.ycombinator.com/item?id=49928868", "by": "gchamonlive", "date": "2026-10-02", "version": null, "stance": "warn", "quote": "settle for such low standards of user experience it's almost pitiful", "excerpt": "I use oh-my-pi, not sure how it compares, but I say someone praising antigravity for being a good harness(1), and for the love of good people settle for such low standards of user experience it's almost pitiful. (1) https://news.ycombinator.com/item?id=49913854", "context": "", "use_case": "AI coding harness", "switched_from": "", "story": "Pi 1.0"}, {"id": 798723008, "item_id": 49920188, "url": "https://news.ycombinator.com/item?id=49920188", "by": "smartbit", "date": "2026-10-01", "version": null, "stance": "warn", "quote": "still many prompts making getting-work-done impossible", "excerpt": "do you mean? accept-edits also available with shift-tab in agy cli [0]. That is not related to executing commands, only to allowing agy edit files. Unless you set Turbo-mode == yolo-mode, agy gui prompts a zillion times too. With 'some features added' I'm referencing this but don't see change in daily work (still many prompts making getting-work-done impossible): v2.14.0 (September 15, 2026) \"New Permissions System\" ", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798670112, "item_id": 49916882, "url": "https://news.ycombinator.com/item?id=49916882", "by": "thanhhaimai", "date": "2026-10-01", "version": null, "stance": "neutral", "quote": "Skills are available with the launch of Antigravity 2", "excerpt": "Opinions are my own. I'm not sure this list is correct. Number 4 is especially wrong, since Skills are available with the launch of Antigravity 2: https://antigravity.google/blog/introducing-google-antigravity-2", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798668096, "item_id": 49916756, "url": "https://news.ycombinator.com/item?id=49916756", "by": "trevorm4", "date": "2026-10-01", "version": null, "stance": "warn", "quote": "", "excerpt": "It has both skills (https://antigravity.google/docs/skills/) and agents (https://antigravity.google/docs/subagents/)", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798649313, "item_id": 49915582, "url": "https://news.ycombinator.com/item?id=49915582", "by": "danpalmer", "date": "2026-09-30", "version": null, "stance": "neutral", "quote": "", "excerpt": "3.8 Flash is my daily driver and produces pretty excellent results all round.", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798642048, "item_id": 49915128, "url": "https://news.ycombinator.com/item?id=49915128", "by": "moecables", "date": "2026-09-30", "version": null, "stance": "neutral", "quote": "I use Antigravity", "excerpt": "I use Antigravity but for some reason, `agy` in the command line feels very bad/incapable of doing things. I can't quite explain it but the most common issue I run into it is just hanging on being unable to finish a tool call", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798623504, "item_id": 49913969, "url": "https://news.ycombinator.com/item?id=49913969", "by": "drusepth", "date": "2026-09-30", "version": null, "stance": "recommend", "quote": "makes me yearn for agy again", "excerpt": "What basic features are missing from agy? I've been using it and cli-cc + web-cc for months (among a few other random harnesses to test here and there) and they all seem roughly comparable to me. I actually just cancelled Ultra also because I couldn't subscribe to a YouTube Family plan while I had it active (Google... :[) but trying to use Codex as a replacement while I testdrive Astra makes me yearn for agy again.", "context": "", "use_case": "coding agent harness", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 798621665, "item_id": 49913854, "url": "https://news.ycombinator.com/item?id=49913854", "by": "spankalee", "date": "2026-09-30", "version": null, "stance": "recommend", "quote": "so is the Antigravity harness", "excerpt": "3.8 Flash is just quite good, and so is the Antigravity harness. I use a mix of Fable 5.1, Opus 5.5, and Gemini 3.8 Flash and Gemini holds it's own. Especially in writing, frontend, and sysadmin work. agy for configuring a NixOS system has been truly incredible.", "context": "", "use_case": "coding harness", "switched_from": "", "story": "Gemini 4 Argon"}, {"id": 797646545, "item_id": 49852909, "url": "https://news.ycombinator.com/item?id=49852909", "by": "smnscu", "date": "2026-09-26", "version": null, "stance": "warn", "quote": "its precision and recall are abysmal", "excerpt": "I do something similar but a bit more involved, using a few informal stages. Let's say for example I'm trying to launch a new complex feature for https://coderba.se. - strategy document - \"sprint\" document with technical implementation - actual implementation - e2e testing scenarios updates Every step involves iterating with Claude on it with me in the loop (setting the direction then resolving the \"founder questions", "context": "", "use_case": "", "switched_from": "", "story": "Plan mode is dead"}, {"id": 796692977, "item_id": 49793311, "url": "https://news.ycombinator.com/item?id=49793311", "by": "rakyll", "date": "2026-09-21", "version": null, "stance": "neutral", "quote": "We use Antigravity for a few generative features", "excerpt": "Disclaimer: I'm one of the co-creators of this project. AX is a layer that is closer to job orchestration, Agent Substrate. It's NOT an agentic framework. We use Antigravity for a few generative features but are abstracting away some of these components so anyone can bring their own. AX is trying to solve some tedious things everyone has to deal with. Layering execution with the underlying stateful worker, wiring up ", "context": "co-creator of AX", "use_case": "generative features", "switched_from": "", "story": "AX \u2013 Google\u2019s Open Agentic Orchestrator"}, {"id": 796502531, "item_id": 49781408, "url": "https://news.ycombinator.com/item?id=49781408", "by": "DanMcInerney", "date": "2026-09-20", "version": null, "stance": "neutral", "quote": "work with any harness like Claude Code, Codex, Antigravity", "excerpt": "I really don't think any of these SOTA labs are doing agentic engineering correctly. Skills are the universal language of all agent harnesses. If you abstract the taste and prescription out of the skills and into guidance docs, then leave the skills as basically just workflow scaffolding, you can build task-specific workflows that work with any harness like Claude Code, Codex, Antigravity, etc. Technically, you only ", "context": "", "use_case": "agent harness", "switched_from": "", "story": "AX \u2013 Google\u2019s Open Agentic Orchestrator"}, {"id": 796397250, "item_id": 49774828, "url": "https://news.ycombinator.com/item?id=49774828", "by": "JoyenBenitto", "date": "2026-09-20", "version": null, "stance": "neutral", "quote": "I have used codex and antigravity but not a huge fan of them.", "excerpt": "Yeah I have been using it to automate almost everything, even my gitlab ci's sometimes when those ci fails I just ask claude to use glab cli to figure things out on it's own. I have used codex and antigravity but not a huge fan of them. I feel my I am too much in favour of claude also because of plugins like GSD and superpower which I am accustomed too.", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: What happens when Claude Code goes away?"}, {"id": 795882150, "item_id": 49742634, "url": "https://news.ycombinator.com/item?id=49742634", "by": "vegadw", "date": "2026-09-17", "version": null, "stance": "recommend", "quote": "On Antigravity/Gemini, I can give it a big list-o-TODOs and have it make a very detailed implementation plan itself, carry it out, and generally do the thing, all unsupervised.", "excerpt": "Not really? I have some custom skills/MCP servers I use: one MCP server for using https://usefeyn.com/blog/pulpie-pareto-optimal-models-for-cleaning-the-web/ ran locally so that reading web pages is cheaper and uses up less context One MCP server for doing automated code quality checking, using Valknut, Jscpd, and Lizard to output scores, combine/aggregate them. That's then give to the model so that it can see if the", "context": "", "use_case": "AI coding agent", "switched_from": "", "story": "Show HN: Share your AI Setup, Learn from others"}, {"id": 795552320, "item_id": 49722020, "url": "https://news.ycombinator.com/item?id=49722020", "by": "qingcharles", "date": "2026-09-16", "version": null, "stance": "recommend", "quote": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me.", "excerpt": "I've used Antigravity as my main coding agent on one of my biggest projects for about a year. It's been great for me. (and I use Claude, Codex, Grok and Muse for all the other projects)", "context": "", "use_case": "main coding agent", "switched_from": "", "story": "Gemini 3.8 Live and 3.8 Live Extended Thinking"}, {"id": 794771825, "item_id": 49673239, "url": "https://news.ycombinator.com/item?id=49673239", "by": "vallerie", "date": "2026-09-12", "version": null, "stance": "recommend", "quote": "I've found good success with the new Gemini models on Antigravity.", "excerpt": "I've found good success with the new Gemini models on Antigravity. Granted I use my models either: - like a fancy auto complete (here are some stub methods, they should do X, fill them in) - using fairly detailed plans and test harnesses, so blowing up the world is hard The 3.X Flash family have been fairly capable models, and the selling point for me is just raw speed. Gemini is noticeably faster than the competitio", "context": "", "use_case": "coding", "switched_from": "", "story": "Ask HN: What default model do you use and why?"}, {"id": 792808688, "item_id": 49550543, "url": "https://news.ycombinator.com/item?id=49550543", "by": "deno", "date": "2026-09-03", "version": null, "stance": "warn", "quote": "still you won't be able to use even the free tier of Antigravity.", "excerpt": "Also word of warning: don't buy Google AI subscription before you confirm you have \"Antigravity entitlement\" whatever that is. You can have \"verified\" account (for 18+ and Gmail), bank card added, active AI Pro subscription, and still you won't be able to use even the free tier of Antigravity. And their verification is some major AI slop that just doesn't work. This is open from June but this was already a bug in Gem", "context": "", "use_case": "", "switched_from": "", "story": "Google Antigravity TOS: 3rd party usage can get Google account suspended"}, {"id": 792794016, "item_id": 49549626, "url": "https://news.ycombinator.com/item?id=49549626", "by": "handzhiev", "date": "2026-09-03", "version": null, "stance": "recommend", "quote": "AGY is best probably, yes. That's what I use.", "excerpt": "AGY is best probably, yes. That's what I use. It works with others too, like OpenCode etc.", "context": "", "use_case": "agent harness for Gemini", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792650080, "item_id": 49540630, "url": "https://news.ycombinator.com/item?id=49540630", "by": "drusepth", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "Antigravity is probably the best of the bunch I've tried.", "excerpt": "Antigravity is probably the best of the bunch I've tried. I'd say it's pretty comparable to Claude Code (I use both daily).", "context": "", "use_case": "harness for Gemini models", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792623552, "item_id": 49538972, "url": "https://news.ycombinator.com/item?id=49538972", "by": "porridgeraisin", "date": "2026-09-02", "version": null, "stance": "recommend", "quote": "agy is good for those cases where you are willing to put the effort into the harness", "excerpt": "agy is good for those cases where you are willing to put the effort into the harness specifically for a task or family of tasks. The full suite, with evals, monitoring, hooks, custom tools, custom verifiers, etc,. It is not good if you want a \"general coding assistant\" like codex or claudecode. The reality is that if you optimise a harness for a family of tasks[1], then most of these models give successful output. An", "context": "", "use_case": "", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792294016, "item_id": 49518376, "url": "https://news.ycombinator.com/item?id=49518376", "by": "sanarothe", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "The codebot produces everything I ask", "excerpt": "Also curious to see what issues people are having. I'm a not-software engineer but I use antigravity quite a bit for my own projects. If you give it a task, it completes it. It might so a single agent chug, or delegate some parallel tasks.. The architecture is essentially what was agreed on during the planning and spec phases. So I'm just not understanding like... The codebot produces everything I ask, I use Gemini a", "context": "not a software engineer", "use_case": "personal projects", "switched_from": "", "story": "Google Antigravity introduces Boost deep reasoning (/boost)"}, {"id": 790646352, "item_id": 49415397, "url": "https://news.ycombinator.com/item?id=49415397", "by": "DefamationStati", "date": "2026-08-24", "version": null, "stance": "warn", "quote": "the limits as absolutely abysmal when using 3.7 flash in antigravity", "excerpt": "I've got the AI pro plan for Gemini and the limits as absolutely abysmal when using 3.7 flash in antigravity, If I use swarms then I can't even finish a single prompt without it reaching the limit.", "context": "has Gemini AI Pro plan", "use_case": "AI coding", "switched_from": "", "story": "Anthropic's best AI model struggles to attract users as cheaper tools thrive"}]}