{"id":96043,"date":"2026-04-09T13:13:18","date_gmt":"2026-04-09T10:13:18","guid":{"rendered":"https:\/\/forklog.com\/en\/?p=96043"},"modified":"2026-04-09T13:15:12","modified_gmt":"2026-04-09T10:15:12","slug":"anthropic-launches-platform-for-ai-agent-deployment","status":"publish","type":"post","link":"https:\/\/forklog.com\/en\/anthropic-launches-platform-for-ai-agent-deployment\/","title":{"rendered":"Anthropic Launches Platform for AI Agent Deployment"},"content":{"rendered":"<p>AI startup Anthropic has introduced an environment for executing complex and prolonged agent tasks\u2014Claude Managed Agents.<\/p>\n<blockquote class=\"twitter-tweet\">\n<p lang=\"en\" dir=\"ltr\">New on the Engineering Blog: <\/p>\n<p>Building Managed Agents\u2014our hosted service for long-running agents\u2014meant solving an old problem in computing: how to design a system for \u201cprograms as yet unthought of.\u201d<\/p>\n<p>Read more: <a href=\"https:\/\/t.co\/YYaEub2QGV\">https:\/\/t.co\/YYaEub2QGV<\/a><\/p>\n<p>\u2014 Anthropic (@AnthropicAI) <a href=\"https:\/\/twitter.com\/AnthropicAI\/status\/2041929199976640948?ref_src=twsrc%5Etfw\">April 8, 2026<\/a><\/p><\/blockquote>\n<p> <script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<p>The solution is a hosting service on the Claude platform, enabling the management of long-term agents through a set of basic interfaces.<\/p>\n<p>The system can sequentially execute tasks, use tools, run code, edit files, interact with external services, and continue functioning even after failures.<\/p>\n<h2 class=\"wp-block-heading\">The Problem<\/h2>\n<p>Previously, developers faced challenges when deploying AI agents:<\/p>\n<ul class=\"wp-block-list\">\n<li>digital assistants would forget context, necessitating a reset;<\/li>\n<li>the model would prematurely end a task, requiring special &#8220;workarounds&#8221;;<\/li>\n<li>the neural network struggled with lengthy assignments.<\/li>\n<\/ul>\n<p>For instance, Claude Sonnet 4.5 could prematurely conclude a task as the context window limit approached.<\/p>\n<h2 class=\"wp-block-heading\">The Solution<\/h2>\n<p>The cloud-based tool Claude Managed Agents handles all background processes, including memory management and error recovery.<\/p>\n<p>The main architectural concept of the new solution is the division of the agent into independent components. Previously, everything operated in one place:<\/p>\n<ul class=\"wp-block-list\">\n<li>the model itself and the logic of its invocation;<\/li>\n<li>code execution;<\/li>\n<li>memory and session states;<\/li>\n<li>accesses and tokens.<\/li>\n<\/ul>\n<p>Anthropic divided the system into three main blocks:<\/p>\n<ol class=\"wp-block-list\">\n<li>Session\u2014a log of all actions and events. It acts as the agent&#8217;s long-term memory, recording the history of steps and bypassing context length limitations.<\/li>\n<li>Control logic\u2014this &#8220;conductor&#8221; calls Claude, decides which context to provide the model, and manages the work cycle.<\/li>\n<li>Execution environment\u2014a place where <span data-descr=\"large language model\" class=\"old_tooltip\">LLM<\/span> can run code, edit files, and interact with the environment.<\/li>\n<\/ol>\n<p>This structure allows each part to function autonomously\u2014a local failure in one of the processes no longer leads to the closure of the entire session.<\/p>\n<p>Anthropic positions Claude Managed Agents as a resilient infrastructure layer that will outlast the evolution of models. It is akin to an operating system for AI agents.<\/p>\n<p>Previously, the company developed a new LLM Claude Mythos, but <a href=\"https:\/\/forklog.com\/en\/news\/anthropic-restricts-public-access-to-ai-model-mythos-after-laboratory-escape\">declined to release<\/a> it publicly due to high security risks.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>AI startup Anthropic has introduced an environment for executing complex and prolonged agent tasks\u2014Claude Managed Agents.<\/p>\n","protected":false},"author":1,"featured_media":96044,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"select":"1","news_style_id":"1","cryptorium_level":"","_short_excerpt_text":"Anthropic unveils Claude Managed Agents for complex AI tasks.","creation_source":"","_metatest_mainpost_news_update":false,"footnotes":""},"categories":[3],"tags":[1751,1434,438],"class_list":["post-96043","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-and-analysis","tag-ai-agents","tag-anthropic","tag-artificial-intelligence"],"aioseo_notices":[],"amp_enabled":true,"views":"43","promo_type":"1","layout_type":"1","short_excerpt":"Anthropic unveils Claude Managed Agents for complex AI tasks.","is_update":"","_links":{"self":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/96043","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/comments?post=96043"}],"version-history":[{"count":1,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/96043\/revisions"}],"predecessor-version":[{"id":96045,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/96043\/revisions\/96045"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media\/96044"}],"wp:attachment":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media?parent=96043"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/categories?post=96043"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/tags?post=96043"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}