--- title: Realtime multiplayer, automation, and you! — Idan Gazit,... description: Idan Gazit, head of GitHub Next, presents two prototypes exploring the future of AI-assisted software development. The first, Agentic Workflows, lets... canonical: https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Realtime multiplayer, automation, and you! — Idan Gazit, GitHub | daily.dev og:description: Idan Gazit, head of GitHub Next, presents two prototypes exploring the future of AI-assisted software development. The first, Agentic Workflows, lets... og:url: https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm og:image: https://api.daily.dev/og/posts/6YONEsJVM.png og:image:alt: Realtime multiplayer, automation, and you! — Idan Gazit, GitHub og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Realtime multiplayer, automation, and you! — Idan Gazit, GitHub **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 21 min read · 2 upvotes · 0 comments ## Summary Idan Gazit, head of GitHub Next, presents two prototypes exploring the future of AI-assisted software development. The first, Agentic Workflows, lets developers write automation playbooks in plain markdown — describing tasks like dependency upgrades in natural language, which Copilot compiles into executable workflows with deterministic security guardrails (scoped permissions, network allowlists, safe output limits). The second, Ace, is a real-time multiplayer development environment where teammates collaborate in a Slack-like interface backed by cloud micro-VMs, with AI acting on the full conversation history to implement decisions. Gazit argues that AI so far has only addressed ~5% of developer work (typing), and the next frontier is helping with the other 95%: understanding codebases, aligning teams, and automating judgment-based tasks. Agentic Workflows is in public preview; Ace is heading to technical preview. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is GitHub's Agentic Workflows and what can it do? Agentic Workflows is a GitHub Next prototype that lets developers write markdown documents describing automation tasks in plain language, which Copilot compiles into GitHub Actions workflows. It was demoed automatically checking for dependency updates, reading changelogs, applying code changes for breaking changes, verifying builds, and opening a single pull request, all restricted by YAML front-matter guardrails specifying permissions, allowed network destinations, and permitted outputs. It is in public preview. _Developers automating dependency upgrades can track how Agentic Workflows evolves on daily.dev._ ### What is GitHub Ace and how is it different from other AI coding tools? Ace is a GitHub Next prototype for real-time multiplayer software development that resembles a Slack-like chat interface where every session runs as a branch of a repository in a cloud micro VM rather than on a local machine. Teammates discuss decisions in chat, and an AI agent reads the full conversation history to infer the final agreed-upon direction and implement it, including acting on collaboratively edited markdown plan documents. It is expected to reach technical preview later in the month announced. _Teams weighing collaborative AI coding workflows can follow Ace's rollout on daily.dev._ ### How does GitHub Agentic Workflows prevent AI agents from being prompt-injected into doing dangerous actions? Agentic Workflows enforces guardrails deterministically in YAML front matter rather than relying on prompt instructions, since prompted guardrails can be overridden by prompt injection. Rules include read-only permissions by default, an allowlist of network destinations the agent may contact (such as npm and the relevant docs site), a safe-outputs block restricting what the agent may write (such as a single pull request only, and an explicit 'do nothing' option), and never exposing secrets to the agent itself, which must request access through a separate warden process. _Engineers designing safe AI agent permissions can dig deeper into these guardrail patterns on daily.dev._ ## Similar posts on daily.dev - [My Experience with GitHub Agentic Workflows](https://daily.dev/posts/my-experience-with-github-agentic-workflows-orst3dlvf) · Thomas Thornton · 0 upvotes · 0 comments - [GitHub Agentic Workflows](https://daily.dev/posts/github-agentic-workflows-6xigduvrg) · Hacker News · 1 upvotes · 0 comments - [A GitHub agentic workflow](https://daily.dev/posts/a-github-agentic-workflow-xqc7skb2g) · A Java geek · 53 upvotes · 2 comments --- Tags: [#github](https://daily.dev/tags/github), [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Realtime multiplayer, automation, and you! — Idan Gazit, GitHub","url":"https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm"},"datePublished":"2026-08-08T21:37:26.336Z","dateModified":"2026-09-14T06:29:23.176Z","description":"Idan Gazit, head of GitHub Next, presents two prototypes exploring the future of AI-assisted software development. The first, Agentic Workflows, lets...","image":"https://i.ytimg.com/vi/iQ5xldZ9StU/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/iQ5xldZ9StU/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"github,ai-agents","timeRequired":"PT21M","video":{"@type":"VideoObject","name":"Realtime multiplayer, automation, and you! — Idan Gazit, GitHub","description":"Idan Gazit, head of GitHub Next, presents two prototypes exploring the future of AI-assisted software development. The first, Agentic Workflows, lets...","thumbnailUrl":"https://i.ytimg.com/vi/iQ5xldZ9StU/sddefault.jpg","uploadDate":"2026-08-08T21:37:26.336Z","duration":"PT21M","url":"https://api.daily.dev/r/6YONEsJVM","embedUrl":"https://www.youtube.com/embed/iQ5xldZ9StU"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"Realtime multiplayer, automation, and you! — Idan Gazit, GitHub"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/realtime-multiplayer-automation-and-you-idan-gazit-github-6yonesjvm#faq","mainEntity":[{"@type":"Question","name":"What is GitHub's Agentic Workflows and what can it do?","acceptedAnswer":{"@type":"Answer","text":"Agentic Workflows is a GitHub Next prototype that lets developers write markdown documents describing automation tasks in plain language, which Copilot compiles into GitHub Actions workflows. It was demoed automatically checking for dependency updates, reading changelogs, applying code changes for breaking changes, verifying builds, and opening a single pull request, all restricted by YAML front-matter guardrails specifying permissions, allowed network destinations, and permitted outputs. It is in public preview. Developers automating dependency upgrades can track how Agentic Workflows evolves on daily.dev."}},{"@type":"Question","name":"What is GitHub Ace and how is it different from other AI coding tools?","acceptedAnswer":{"@type":"Answer","text":"Ace is a GitHub Next prototype for real-time multiplayer software development that resembles a Slack-like chat interface where every session runs as a branch of a repository in a cloud micro VM rather than on a local machine. Teammates discuss decisions in chat, and an AI agent reads the full conversation history to infer the final agreed-upon direction and implement it, including acting on collaboratively edited markdown plan documents. It is expected to reach technical preview later in the month announced. Teams weighing collaborative AI coding workflows can follow Ace's rollout on daily.dev."}},{"@type":"Question","name":"How does GitHub Agentic Workflows prevent AI agents from being prompt-injected into doing dangerous actions?","acceptedAnswer":{"@type":"Answer","text":"Agentic Workflows enforces guardrails deterministically in YAML front matter rather than relying on prompt instructions, since prompted guardrails can be overridden by prompt injection. Rules include read-only permissions by default, an allowlist of network destinations the agent may contact (such as npm and the relevant docs site), a safe-outputs block restricting what the agent may write (such as a single pull request only, and an explicit 'do nothing' option), and never exposing secrets to the agent itself, which must request access through a separate warden process. Engineers designing safe AI agent permissions can dig deeper into these guardrail patterns on daily.dev."}}]} ```