[{"data":1,"prerenderedAt":838},["ShallowReactive",2],{"search-navigation":3,"\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2":36,"\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2-surround":279},[4,8,12,16,20,24,28,32],{"title":5,"path":6,"stem":7},"Solving AI Agent Skills Fragmentation with npx skills","\u002Farticle\u002Fcentralised-the-skills-for-agents","en\u002Farticle\u002Fcentralised-the-skills-for-agents",{"title":9,"path":10,"stem":11},"Stop Exposing Your Vault: Secure NodeWarden with Cloudflare","\u002Farticle\u002Fhow-to-secure-nodewarden-at-cloudflare","en\u002Farticle\u002Fhow-to-secure-nodewarden-at-cloudflare",{"title":13,"path":14,"stem":15},"How to Self-Host a BitWarden Compatible Password Vault","\u002Farticle\u002Fself-hosting-nodewarden-at-cloudflare","en\u002Farticle\u002Fself-hosting-nodewarden-at-cloudflare",{"title":17,"path":18,"stem":19},"Google Antigravity 2 Hands-On: Setup & First Impressions","\u002Farticle\u002Fsetup-antigravity-2","en\u002Farticle\u002Fsetup-antigravity-2",{"title":21,"path":22,"stem":23},"NetSuite AI Connector Guide: MCP Setup & Use Cases","\u002Farticle\u002Fsetup-netsuite-mcp-on-chatgpt-claude","en\u002Farticle\u002Fsetup-netsuite-mcp-on-chatgpt-claude",{"title":25,"path":26,"stem":27},"Testing Claude Fable 5: A Financial Analysis Review","\u002Farticle\u002Ftesting-claude-fable-5-in-financial-analysis","en\u002Farticle\u002Ftesting-claude-fable-5-in-financial-analysis",{"title":29,"path":30,"stem":31},"Guide to Browser and Scheduler in Antigravity 2","\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2","en\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2",{"title":33,"path":34,"stem":35},"Antigravity 2 in Action: AI Agents, Skills, and MCP","\u002Farticle\u002Fwhat-can-we-do-with-antigravity-2","en\u002Farticle\u002Fwhat-can-we-do-with-antigravity-2",{"id":37,"title":29,"author":38,"body":39,"category":38,"date":264,"description":265,"extension":266,"image":267,"imageAlt":268,"isPublished":111,"meta":269,"minRead":275,"navigation":111,"path":276,"seo":277,"stem":31,"__hash__":278},"article\u002Fen\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2.md",null,{"type":40,"value":41,"toc":258},"minimark",[42,60,65,70,80,95,101,104,116,120,123,130,137,148,151,157,168,178,189,204,214,224,227,233,237,248],[43,44,45,46,50,51,54,55,59],"p",{},"The ",[47,48,49],"code",{},"\u002Fbrowser"," and ",[47,52,53],{},"\u002Fschedule"," commands in ",[56,57,58],"strong",{},"Google Antigravity 2"," redefine how we interact with AI. These two commands transform Antigravity into a proactive automation platform that can interact with the live web and run recurring workflows.",[61,62,64],"h2",{"id":63},"browser-run","Browser Run",[43,66,45,67,69],{},[47,68,49],{}," command forces the use of the Chrome browser to complete a task. Users remain in control over when and how agents access the web during the workflow. Screenshots and video clips are taken during the process.",[43,71,72,75],{},[56,73,74],{},"Use Case:",[76,77],"img",{"alt":78,"src":79},"Using the \u002Fbrowser command to submit sick leave","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-command-sick-leave-prompt.webp",[43,81,82,83,87,91],{},"The following are the output from the agent:\n",[76,84],{"alt":85,"src":86},"Browser automation subagent starting the sick leave submission task","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-automation-subagent-delegation.webp",[76,88],{"alt":89,"src":90},"Browser automator setting up Chrome with remote debugging and permissions","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-automation-permissions-and-chrome-setup.webp",[76,92],{"alt":93,"src":94},"Browser subagent pausing at a login wall requesting user credentials","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-automation-login-wall-encountered.webp",[43,96,97],{},[76,98],{"alt":99,"src":100},"Browser subagent waiting for the user to complete the login process","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-automation-waiting-for-user-login.webp",[43,102,103],{},"Antigravity was capable of opening the page and submitting the form. When a login is required, the agent will wait until you have logged in to the website.",[43,105,106,112,113,115],{},[107,108],"video",{"alt":109,"src":110,"controls":111},"Video of the browser subagent executing the sick leave submission process","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fbrowser-automation-sick-leave-process.webm",true,"\nThe ",[47,114,49],{}," command is useful when we need the most recent documentation from the web, change logs, testing web applications, or checking API statuses to complete a task.",[61,117,119],{"id":118},"schedule-task","Schedule Task",[43,121,122],{},"Antigravity 2 has built-in functionality for scheduling tasks. We will test it out by auto-entering a timesheet every 10 minutes.",[43,124,125,126],{},"Click on 'Scheduled Tasks':\n",[76,127],{"alt":128,"src":129},"Scheduled Tasks menu option in the sidebar","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-menu-item.webp",[43,131,132,133],{},"Click on '+ New' button to create a new scheduled task:\n",[76,134],{"alt":135,"src":136},"Empty Scheduled Tasks view with a new task button","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-empty-state-new-button.webp",[43,138,139,140,144],{},"Fill in the form and schedule the task to run every 10 minutes:\n",[76,141],{"alt":142,"src":143},"New Scheduled Task form filled out for auto timesheet entry","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-new-task-form.webp",[76,145],{"alt":146,"src":147},"Active Auto Timesheet Entry task in the Scheduled Tasks list","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-active-task-list.webp",[43,149,150],{},"The task was triggered, but it resulted in an error.",[43,152,153],{},[76,154],{"alt":155,"src":156},"Error message indicating no available models found for tier flash","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-error-no-models.webp",[43,158,159,160,163,164,167],{},"We tried adding 'env' values to the ",[47,161,162],{},"sidecar.json"," file, which is located at ",[47,165,166],{},".gemini\u002Fconfig\u002Fsidecars\u002Fauto-timesheet-entry",", but we still received the same error.",[169,170,175],"pre",{"className":171,"code":173,"language":174},[172],"language-text","    \"env\": {\n        \"GEMINI_MODEL\": \"gemini-3.5-flash\",\n        \"GOOGLE_CLOUD_PROJECT\": \"your-gcp-project-id\"\n    }\n","text",[47,176,173],{"__ignoreMap":177},"",[43,179,180,181,184,185,188],{},"Inspecting the scheduled task logs located at ",[47,182,183],{},".gemini\u002Fantigravity\u002Fsidecar_data\u002F",", we can see ",[47,186,187],{},"[schedule] Triggering command: \"agentapi\" [new-conversation]",".",[43,190,191,192,199,200,203],{},"The response to our issue posted in the ",[193,194,198],"a",{"href":195,"rel":196},"https:\u002F\u002Fdiscuss.ai.google.dev\u002Ft\u002Fscheduled-tasks-in-antigravity-2-ended-with-errors\u002F168013",[197],"nofollow","Build with Google AI Forum"," suggests an API key is required for ",[47,201,202],{},"agentapi",". We tried using the API key, but still had no luck.",[43,205,206,207,209,210],{},"We tried again with the command, ",[47,208,53],{},", directly in the conversation box.\n",[76,211],{"alt":212,"src":213},"Using the \u002Fschedule command to prompt the timesheet automation task","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fschedule-command-timesheet-prompt.webp",[43,215,216,217,219,220],{},"Tasks scheduled using the ",[47,218,53],{}," command will not appear under the 'Scheduled Tasks' menu item.\n",[76,221],{"alt":222,"src":223},"Scheduled Tasks list not showing tasks created via CLI command","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fscheduled-tasks-list-missing-cli-task.webp",[43,225,226],{},"The cron task did get triggered.",[43,228,229],{},[76,230],{"alt":231,"src":232},"Cron job log confirming it was triggered","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002Fcron-job-triggered-log.webp",[61,234,236],{"id":235},"final-thoughts","Final thoughts",[43,238,45,239,241,242,244,245,247],{},[47,240,49],{},"and ",[47,243,53],{}," commands in Antigravity 2 are not just isolated features, combining them creates a powerful tool. For example, scheduling an agent to use ",[47,246,49],{}," to navigate to news and information websites every Monday morning at 6:00 AM, synthesise the info, and deliver a comprehensive daily briefing.",[43,249,250,251,254,255,257],{},"Other commands, such as ",[47,252,253],{},"\u002Fgrill-me"," (mentioned in the article ",[193,256,33],{"href":34},"), can be used when designing new scheduled tasks. It prompts the agent to ask users to clarify ambiguities and catch edge cases before finalising the design.",{"title":177,"searchDepth":259,"depth":259,"links":260},2,[261,262,263],{"id":63,"depth":259,"text":64},{"id":118,"depth":259,"text":119},{"id":235,"depth":259,"text":236},"2026-06-07","Discover how the \u002Fbrowser and \u002Fschedule commands elevate Google Antigravity 2 into a powerful AI automation platform.","md","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FUse%20Browser%20and%20Scheduler%20in%20Antigravity%202\u002FEN-Use-Browser-and-Scheduler-in-Antigravity-2.webp","Google Antigravity 2: Browser & Schedule Automation",{"schemaOrg":270,"robots":273,"sitemap":274},[271],{"@type":272,"headline":268,"description":265},"TechArticle","index, follow",{"lastmod":264},5,"\u002Fen\u002Farticle\u002Fuse-browser-and-scheduler-in-antigravity-2",{"title":268,"description":265,"ogTitle":29,"ogDescription":265,"twitterTitle":29,"twitterDescription":265},"4mcPjN-m89F0bW0uV2vlHDsS_g7gItd9UCcxf8hMcAs",[280,629],{"id":281,"title":33,"author":177,"body":282,"category":177,"date":619,"description":620,"extension":266,"image":621,"imageAlt":622,"isPublished":111,"meta":623,"minRead":275,"navigation":111,"path":34,"seo":627,"stem":35,"__hash__":628},"article\u002Fen\u002Farticle\u002Fwhat-can-we-do-with-antigravity-2.md",{"type":40,"value":283,"toc":608},[284,287,293,297,304,307,311,317,320,326,332,338,350,355,360,368,372,378,384,390,393,399,405,410,415,418,422,432,438,442,448,454,460,463,467,470,474,480,486,492,495,582,588,598,601,604],[43,285,286],{},"Antigravity 2, as an agent-first desktop application, allows users to orchestrate a team of subagents to generate a daily brief from emails and calendars, or to build, test, and deploy an application. But what does this actually look like in practice?",[43,288,289,290,292],{},"Let’s explore what Antigravity 2 can do for you—from built-in commands and skills to how integrations like Firebase Model Context Protocol (MCP) servers are turning developers into full-stack architects. Please refer to this article, ",[193,291,17],{"href":18},", for setting up and configuring skills and MCP servers.",[61,294,296],{"id":295},"built-in-commands","Built-in Commands",[298,299,301],"h3",{"id":300},"goal",[47,302,303],{},"\u002Fgoal",[43,305,306],{},"With this command, you tell the agent to run until a specific task is completely finished.",[43,308,309],{},[56,310,74],{},[43,312,313],{},[76,314],{"alt":315,"src":316},"Prompting the Antigravity 2 goal command to build a Textris game","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-goal-command-prompt.webp",[43,318,319],{},"When we ask for dynamic subagents, three agents are spawned to complete the task.",[43,321,322],{},[76,323],{"alt":324,"src":325},"Antigravity 2 spawning subagents to complete tasks autonomously","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-spawning-subagents-1.webp",[43,327,328],{},[76,329],{"alt":330,"src":331},"Detailed progress of Antigravity 2 subagents executing steps","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-spawning-subagents-2.webp",[43,333,334],{},[76,335],{"alt":336,"src":337},"Completion of autonomous tasks by Antigravity 2 subagents","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-spawning-subagents-3.webp",[43,339,340,341,345,346],{},"To play the generated Textris game in HTML, please click this ",[193,342,344],{"href":343},"\u002Ftextris","link",":\n",[107,347],{"alt":348,"src":349,"controls":111},"AI generated Textris game in HTML","https:\u002F\u002Fpub-9791558bbefc4bb59834a03ae5c44137.r2.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fai-generated-textri-game-html.webm",[43,351,45,352,354],{},[47,353,303],{}," command creates a plan, executes it step by step, and reports the progress to you at each stage. Additionally, the agent will manage the context window itself without exceeding limits. This is suitable for lengthy tasks, such as performance optimisation or module refactoring, which require multiple rounds of iteration.",[298,356,358],{"id":357},"grill-me",[47,359,253],{},[43,361,362,363,367],{},"This command triggers the skill ",[193,364,357],{"href":365,"rel":366},"https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fproductivity\u002Fgrill-me\u002FSKILL.md",[197],", created by Matt Pocock. It is concise and highly effective. The idea behind it is to question you from multiple aspects, clarify edge cases, align the structures, and finalise the plan before implementation.",[43,369,370],{},[56,371,74],{},[43,373,374],{},[76,375],{"alt":376,"src":377},"Triggering the grill-me command for Cloudflare architecture in Antigravity 2","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-grill-me-command-prompt.webp",[43,379,380],{},[76,381],{"alt":382,"src":383},"Interactive questioning session using the grill-me skill","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-grill-me-interaction-1.webp",[43,385,386],{},[76,387],{"alt":388,"src":389},"Clarifying edge cases and structuring the plan with grill-me","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-grill-me-interaction-2.webp",[43,391,392],{},"It supports multi-select capabilities:",[43,394,395],{},[76,396],{"alt":397,"src":398},"Using multi-select capabilities during the grill-me process","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-grill-me-multi-select-1.webp",[43,400,401],{},[76,402],{"alt":403,"src":404},"Finalizing choices with the grill-me multi-select feature","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-grill-me-multi-select-2.webp",[43,406,45,407,409],{},[47,408,253],{}," command is quite impressive. It covers almost all aspects of architectural design for deploying to Cloudflare. This makes it suitable for the architectural design of an app, refactoring across multiple domains, and other potentially complex tasks.",[298,411,412],{"id":177},[47,413,414],{},"@\u002F ",[43,416,417],{},"This command doesn't appear on the available list, but we sometimes need it for our context window. It is used to add a folder or files (other than media) to the conversation box.",[61,419,421],{"id":420},"skills","Skills",[43,423,424,425,428,429,188],{},"We can type ",[47,426,427],{},"\u002F"," to activate the popup box and select a skill. Let's test out ",[47,430,431],{},"excalidraw-diagram",[43,433,434],{},[76,435],{"alt":436,"src":437},"Selecting the Excalidraw diagram skill from the Antigravity popup box","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-skill-selection-excalidraw.webp",[43,439,440],{},[56,441,74],{},[43,443,444],{},[76,445],{"alt":446,"src":447},"Providing instructions to generate an Excalidraw diagram","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-excalidraw-diagram-prompt.webp",[43,449,450],{},[76,451],{"alt":452,"src":453},"Successful generation of an Excalidraw architecture diagram","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-excalidraw-diagram-generation.webp",[43,455,456],{},[76,457],{"alt":458,"src":459},"Antigravity architecture diagram","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-architecture.webp",[43,461,462],{},"The skill was successfully executed after a couple of rounds of permission confirmations.",[61,464,466],{"id":465},"mcp-servers","MCP Servers",[43,468,469],{},"When the agent builds a backend, you usually have to open a web console, create a database, copy API keys, and paste them back. By leveraging MCP servers directly in Antigravity, the agent gains \"hands\" to manipulate your cloud infrastructure.",[43,471,472],{},[56,473,74],{},[43,475,476],{},[76,477],{"alt":478,"src":479},"Instructing the MCP server to create a Firebase web application","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-mcp-server-firebase-prompt.webp",[43,481,482],{},[76,483],{"alt":484,"src":485},"Antigravity using MCP servers to manipulate Firebase infrastructure","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-mcp-server-firebase-execution-1.webp",[43,487,488],{},[76,489],{"alt":490,"src":491},"Successful creation of a Firebase app via Antigravity MCP servers","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-mcp-server-firebase-execution-2.webp",[43,493,494],{},"The agent successfully managed to create the web app in Firebase using MCP servers. Let's check the repository.",[169,496,500],{"className":497,"code":498,"language":499,"meta":177,"style":177},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","Create a new public GitHub repository called ‘to-do-app-techvoyage’. Then push all the current project files to it with the commit message ‘Initial commit — to-do-app’.\n","bash",[47,501,502],{"__ignoreMap":177},[503,504,507,511,515,518,521,524,527,530,533,536,539,542,545,548,551,554,557,560,563,565,568,571,574,576,579],"span",{"class":505,"line":506},"line",1,[503,508,510],{"class":509},"sBMFI","Create",[503,512,514],{"class":513},"sfazB"," a",[503,516,517],{"class":513}," new",[503,519,520],{"class":513}," public",[503,522,523],{"class":513}," GitHub",[503,525,526],{"class":513}," repository",[503,528,529],{"class":513}," called",[503,531,532],{"class":513}," ‘to-do-app-techvoyage’.",[503,534,535],{"class":513}," Then",[503,537,538],{"class":513}," push",[503,540,541],{"class":513}," all",[503,543,544],{"class":513}," the",[503,546,547],{"class":513}," current",[503,549,550],{"class":513}," project",[503,552,553],{"class":513}," files",[503,555,556],{"class":513}," to",[503,558,559],{"class":513}," it",[503,561,562],{"class":513}," with",[503,564,544],{"class":513},[503,566,567],{"class":513}," commit",[503,569,570],{"class":513}," message",[503,572,573],{"class":513}," ‘Initial",[503,575,567],{"class":513},[503,577,578],{"class":513}," —",[503,580,581],{"class":513}," to-do-app’.\n",[43,583,584],{},[76,585],{"alt":586,"src":587},"Agent automatically pushing project files to a new GitHub repository","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002Fantigravity-github-repository-push.webp",[43,589,45,590,593,594,188],{},[47,591,592],{},"to-do-app-techvoyage"," repository is available on GitHub via this ",[193,595,344],{"href":596,"rel":597},"https:\u002F\u002Fgithub.com\u002FTech-Voyage-Dev\u002Fto-do-app-techvoyage",[197],[61,599,600],{"id":235},"Final Thoughts",[43,602,603],{},"While MCP and Skills in Antigravity 2—powered by Gemini 3.5 Flash—deliver blazing speed, the agentic flow is currently bottlenecked by constant permission interruptions and a lack of checkpoints for rollbacks. Until the checkpoint feature is introduced, frequent Git commits are your mandatory safety net.",[605,606,607],"style",{},"html pre.shiki code .sBMFI, html code.shiki .sBMFI{--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B}html pre.shiki code .sfazB, html code.shiki .sfazB{--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D}html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"title":177,"searchDepth":259,"depth":259,"links":609},[610,616,617,618],{"id":295,"depth":259,"text":296,"children":611},[612,614,615],{"id":300,"depth":613,"text":303},3,{"id":357,"depth":613,"text":253},{"id":177,"depth":613,"text":414},{"id":420,"depth":259,"text":421},{"id":465,"depth":259,"text":466},{"id":235,"depth":259,"text":600},"2026-06-06","Explore Antigravity 2, an agent-first desktop app. Learn to orchestrate AI subagents, use built-in skills, and leverage MCP servers to build full-stack apps.","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FWhat%20can%20we%20do%20with%20Antigravity%202\u002FEN-What-can-we-do-with-Antigravity-2.webp","Antigravity 2 Tutorial: AI Agents, Skills & MCP Servers",{"schemaOrg":624,"robots":273,"sitemap":626},[625],{"@type":272,"headline":33,"description":620},{"lastmod":619},{"title":622,"description":620,"ogTitle":33,"ogDescription":620,"twitterTitle":33,"twitterDescription":620},"JeSkC_527Foa9wnhH3eb2fNNcILJzNwlr9BteGQ5eIg",{"id":630,"title":25,"author":38,"body":631,"category":38,"date":824,"description":825,"extension":266,"image":826,"imageAlt":827,"isPublished":111,"meta":828,"minRead":275,"navigation":111,"path":26,"seo":836,"stem":27,"__hash__":837},"article\u002Fen\u002Farticle\u002Ftesting-claude-fable-5-in-financial-analysis.md",{"type":40,"value":632,"toc":822},[633,644,655,658,664,680,690,696,702,708,730,739,753,761,780,783,794,803,806,819],[43,634,635,636,639,640],{},"On June 9, 2026, Anthropic released ",[56,637,638],{},"Claude Fable 5",", a Mythos-Class model with advanced safeguards in some areas such as cybersecurity, distillation, chemistry and biology.\n",[76,641],{"alt":642,"src":643},"Anthropic Claude Fable 5 launch announcement and model capabilities","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-launch-announcement.webp",[43,645,646,647,650,651],{},"Anthropic has published numbers showing Fable 5 is ahead of Opus 4.8 and GPT-5.5 on several key benchmarks. In ",[56,648,649],{},"GDPval-AA (knowledge work)",", it is significantly ahead of Opus 4.8, GPT 5.5 and Gemini 3.1 Pro.\n",[76,652],{"alt":653,"src":654},"Claude Fable 5 benchmark comparison including GDPval-AA knowledge work score","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-benchmarks-gdpval-aa.webp",[43,656,657],{},"According to Claude, \"Fable 5 shows strong performance on complex analytical tasks.\".",[43,659,660],{},[76,661],{"alt":662,"src":663},"Claude Fable 5 performance statement on complex analytical tasks","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-analytical-performance-statement.webp",[43,665,666,667,670,671,674,675,188],{},"We decided to test ",[56,668,669],{},"Fable 5"," by running an analysis of 'Space Exploration Technologies Corporation(SpaceX)' for investment research purposes, using the '",[56,672,673],{},"comps-analysis","' skill. The 'skill.md' file was from ",[193,676,679],{"href":677,"rel":678},"https:\u002F\u002Fgithub.com\u002Fanthropics\u002Ffinancial-services\u002Fblob\u002Fmain\u002Fplugins\u002Fagent-plugins\u002Fmarket-researcher\u002Fskills\u002Fcomps-analysis\u002FSKILL.md",[197],"Anthropic's financial-services GitHub repository",[43,681,682,686],{},[76,683],{"alt":684,"src":685},"Initiating SpaceX financial comps analysis skill in Claude chat","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fspacex-comps-analysis-skill-initiation.webp",[76,687],{"alt":688,"src":689},"Claude Fable 5 executing automated financial comps analysis for SpaceX","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fspacex-comps-analysis-running-progress.webp",[43,691,692],{},[76,693],{"alt":694,"src":695},"SpaceX automated financial comps analysis workflow final output","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fspacex-comps-analysis-skill-results-terminal.webp",[43,697,698,699,701],{},"The task took around 25 mins. Claude did not ask any clarifying questions except to request permission to access external websites. Impressively, ",[56,700,669],{}," was capable of finding the peer companies for comparison.",[43,703,704],{},[76,705],{"alt":706,"src":707},"Claude Fable 5 requesting explicit permission for external website access","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-external-site-access-permission.webp",[43,709,710,714,715,718,719,724,725,188],{},[76,711],{"alt":712,"src":713},"SpaceX peer group companies identified autonomously by Claude Fable 5","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-spacex-comps-excel-memo.webp","\n*",[56,716,717],{},"Please note:"," This is for research and educational purpose and does not constitute investment advice. Original generated files can be downloaded here, ",[193,720,723],{"href":721,"rel":722},"https:\u002F\u002Ftechvoyage.dev\u002FSpaceX_Comps_Analysis_2026-06-10.xlsx",[197],"SpaceX_Comps_Analysis_2026-06-10.xlsx",", and, ",[193,726,729],{"href":727,"rel":728},"https:\u002F\u002Ftechvoyage.dev\u002FSpaceX_Comps_Memo_2026-06-10.pdf",[197],"SpaceX_Comps_Memo_2026-06-10.pdf",[43,731,732,733,736,737,188],{},"When we tested the same skill using ",[56,734,735],{},"Qwen deep research",", we had to manually enter the peer companies. The numerical figures are differed slightly compared to ",[56,738,669],{},[43,740,741,745,749],{},[76,742],{"alt":743,"src":744},"Initiating SpaceX investment comps analysis in Qwen deep research","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fqwen-deep-research-spacex-comps-initiation.webp",[76,746],{"alt":747,"src":748},"Entering peer companies manually in Qwen deep research for SpaceX analysis","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fqwen-deep-research-spacex-comps-running.webp",[76,750],{"alt":751,"src":752},"SpaceX financial benchmark analysis report preview generated by Qwen","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fqwen-deep-research-spacex-comps-report-preview.webp",[43,754,755,756,188],{},"*The Qwen-generated report can be downloaded here, ",[193,757,760],{"href":758,"rel":759},"https:\u002F\u002Ftechvoyage.dev\u002FThe_SpaceX_Comp_Analysis_A_Statistical_Benchmark_of_Growth_Efficiency_and_Valuation_Against_Public_Peers.pdf",[197],"The_SpaceX_Comp_Analysis_A_Statistical_Benchmark_of_Growth_Efficiency_and_Valuation_Against_Public_Peers.pdf",[43,762,763,764,767,768,771,772,776],{},"Although both reports from Claude and Qwen were produced based on the same methodology specified in the skill, the numbers varied because Claude and Qwen pulled from different sources. For example, Claude Fable 5 used ",[56,765,766],{},"Stock Analysis"," while Qwen used ",[56,769,770],{},"SEC.gov",".\n",[76,773],{"alt":774,"src":775},"Financial data extracted from Stock Analysis by Claude Fable 5","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-data-source-stock-analysis.webp",[76,777],{"alt":778,"src":779},"SEC filings and SEC.gov data source utilized by Qwen deep research","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fqwen-deep-research-data-source-sec-gov.webp",[43,781,782],{},"The report from Qwen also contained much more information regarding the business model, capital structure and operation.",[43,784,785,786,789,790],{},"After comparing reports, we have decided to try again with ",[47,787,788],{},"\u002Fresearch"," command in Claude desktop app.\n",[76,791],{"alt":792,"src":793},"Using the new research command in Claude desktop app for company analysis","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-desktop-research-command-spacex.webp",[43,795,796,797,802],{},"The newly generated analysis report, ",[193,798,801],{"href":799,"rel":800},"https:\u002F\u002Ftechvoyage.dev\u002FSpaceX_Analysis.pdf",[197],"SpaceX_Analysis.pdf",", included extra recommendations and caveats sections. Compared to Qwen deep search, it still lacked detailed company information such as operation and capital structure.",[43,804,805],{},"Is Fable 5 worth the premium price tag?  It might be for hard, long tasks and autonomous coding. But for certain knowledge work, like company analysis, it may not be the best value yet.",[43,807,808,809,811,812,771,815],{},"At then end,  we tested how well ",[56,810,669],{}," performs in translation. Strangely, the model auto switched back to ",[56,813,814],{},"Opus 4.8",[76,816],{"alt":817,"src":818},"Claude desktop app automatically switching model from Fable 5 to Opus 4.8","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002Fclaude-fable-5-translation-opus-switching.webp",[43,820,821],{},"Does this mean any long-context conversion will be reverted back to Opus 4.8 ? Is this just a temporary quirk during the usage before 22nd June? Or is the system simply smart enough to tell the translation doesn't require such a powerful model.",{"title":177,"searchDepth":259,"depth":259,"links":823},[],"2026-06-11","We test Anthropic's new Claude Fable 5 against Qwen Deep Research for financial analysis. Is it worth the upgrade?","https:\u002F\u002Finimages.techvoyage.dev\u002Farticles\u002FTesting%20Claude%20Fable%205%20in%20Financial%20Analysis\u002FEN-Testing-Claude-Fable-5-in-Financial-Analysis.webp","Claude Fable 5 vs. Qwen: AI Financial Analysis Test",{"tags":829,"schemaOrg":833,"robots":273,"sitemap":835},[638,830,831,832],"Financial Analysis","Qwen Deep Research","AI Benchmarks",[834],{"@type":272,"headline":25,"description":825},{"lastmod":824},{"title":827,"description":825,"ogTitle":25,"ogDescription":825,"twitterTitle":25,"twitterDescription":825},"7IMijVENJ7UyveGCrTcSxcOnrnYxXKMuiNS08r03S6s",1781435111037]