--- title: AWS launches open-source AI agent sandbox to prevent... description: AWS released Strands Box, an open-source sandbox for AI agents that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful... canonical: https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: AWS launches open-source AI agent sandbox to prevent YOLO mode disasters | daily.dev og:description: AWS released Strands Box, an open-source sandbox for AI agents that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful... og:url: https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj og:image: https://api.daily.dev/og/posts/PgayLBuCj.png og:image:alt: AWS launches open-source AI agent sandbox to prevent YOLO mode disasters 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. # AWS launches open-source AI agent sandbox to prevent YOLO mode disasters **[The Register](https://daily.dev/sources/theregister)** · 4 min read · 0 upvotes · 0 comments ## Summary AWS released Strands Box, an open-source sandbox for AI agents that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful rules on agent tool calls—such as rate-limiting Slack posts, controlling Git pushes, or capping API spend. Unlike standard containers or microVMs, Box tracks an agent's prior actions and enforces deterministic policies that agents cannot talk their way around, though AWS stresses developers remain responsible for configuring access correctly. Strands Box also bundles Strands Shell and Monty for Python to expose shell and Python operations to the same policy engine. It's available on GitHub now, currently macOS-only, with Linux and Windows support planned alongside deployment targets like AgentCore, ECS, and Kubernetes. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is AWS Strands Box and how does it control AI agent behavior? Strands Box is an open-source sandbox from AWS that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful rules on AI agent tool calls. It tracks what an agent has already done, not just what it's requesting, so policies can limit things like Slack posts to three per ten minutes, restrict Git pushes, or cap API spend. It enforces rules deterministically, without relying on the agent to follow instructions. _daily.dev surfaces tooling like this for developers locking down agentic workflows before shipping them._ ### What platforms does AWS Strands Box currently support? Strands Box is available on GitHub now but only supports macOS at launch. Linux support is in development, and a Windows client is described by AWS as being on its radar, though neither has a planned release date. Deployment to platforms such as AgentCore, ECS, and Kubernetes is also planned but not yet available. _Developers evaluating new agent-safety tooling can track rollout details like this through daily.dev._ ### How is Strands Box different from using containers or microVMs to isolate AI agents? Containers and microVMs provide strong access isolation but lack contextual rule enforcement, meaning once an agent has a tool it can do anything with it, including deleting a production database. Strands Box adds the Dogwood Local Engine for temporal awareness, letting the policy engine check tool calls against prior agent activity, not just the access boundary, closing that gap. _daily.dev helps developers comparing agent isolation approaches keep up with new safety mechanisms._ ## Similar posts on daily.dev - [AWS creates a sandbox for its agent experiments](https://daily.dev/posts/aws-creates-a-sandbox-for-its-agent-experiments-r5p1skjmm) · The New Stack · 0 upvotes · 0 comments - [AWS Launches Strands Labs for Experimental AI Agent Projects](https://daily.dev/posts/aws-launches-strands-labs-for-experimental-ai-agent-projects-h0ycai6cx) · InfoQ · 1 upvotes · 0 comments - [Build and Run Your Own AI Agent in the Cloud](https://daily.dev/posts/build-and-run-your-own-ai-agent-in-the-cloud-aoostzrec) · Towards Data Science · 1 upvotes · 0 comments --- Tags: [#open-source](https://daily.dev/tags/open-source), [#aws](https://daily.dev/tags/aws), [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-security](https://daily.dev/tags/ai-security) [View this post on daily.dev](https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj) ```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":"AWS launches open-source AI agent sandbox to prevent YOLO mode disasters","url":"https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj"},"datePublished":"2026-10-07T17:42:59.922Z","dateModified":"2026-10-07T18:21:28.200Z","description":"AWS released Strands Box, an open-source sandbox for AI agents that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/68cc8b1d173ed842d64195917f0f5629?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/68cc8b1d173ed842d64195917f0f5629?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The Register","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":"The Register","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/66aa2113fdad463992ffcbf0e8963fda","url":"https://daily.dev/sources/theregister"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"open-source,aws,ai-agents,ai-security","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Register","item":"https://daily.dev/sources/theregister"},{"@type":"ListItem","position":3,"name":"AWS launches open-source AI agent sandbox to prevent YOLO mode disasters"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/aws-launches-open-source-ai-agent-sandbox-to-prevent-yolo-mode-disasters-pgaylbucj#faq","mainEntity":[{"@type":"Question","name":"What is AWS Strands Box and how does it control AI agent behavior?","acceptedAnswer":{"@type":"Answer","text":"Strands Box is an open-source sandbox from AWS that combines OS-level isolation with the Dogwood policy engine to enforce contextual, stateful rules on AI agent tool calls. It tracks what an agent has already done, not just what it's requesting, so policies can limit things like Slack posts to three per ten minutes, restrict Git pushes, or cap API spend. It enforces rules deterministically, without relying on the agent to follow instructions. daily.dev surfaces tooling like this for developers locking down agentic workflows before shipping them."}},{"@type":"Question","name":"What platforms does AWS Strands Box currently support?","acceptedAnswer":{"@type":"Answer","text":"Strands Box is available on GitHub now but only supports macOS at launch. Linux support is in development, and a Windows client is described by AWS as being on its radar, though neither has a planned release date. Deployment to platforms such as AgentCore, ECS, and Kubernetes is also planned but not yet available. Developers evaluating new agent-safety tooling can track rollout details like this through daily.dev."}},{"@type":"Question","name":"How is Strands Box different from using containers or microVMs to isolate AI agents?","acceptedAnswer":{"@type":"Answer","text":"Containers and microVMs provide strong access isolation but lack contextual rule enforcement, meaning once an agent has a tool it can do anything with it, including deleting a production database. Strands Box adds the Dogwood Local Engine for temporal awareness, letting the policy engine check tool calls against prior agent activity, not just the access boundary, closing that gap. daily.dev helps developers comparing agent isolation approaches keep up with new safety mechanisms."}}]} ```