{"id":1260,"date":"2025-10-01T06:57:12","date_gmt":"2025-10-01T13:57:12","guid":{"rendered":"https:\/\/devblogs.microsoft.com\/foundry\/?p=1260"},"modified":"2025-10-14T21:42:47","modified_gmt":"2025-10-15T04:42:47","slug":"introducing-microsoft-agent-framework-the-open-source-engine-for-agentic-ai-apps","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/foundry\/introducing-microsoft-agent-framework-the-open-source-engine-for-agentic-ai-apps\/","title":{"rendered":"Introducing Microsoft Agent Framework: The Open-Source Engine for Agentic AI Apps"},"content":{"rendered":"<h2>Why agents need a new foundation<\/h2>\n<p>Over the last year, developers have been experimenting with <strong>AI agents<\/strong> in every imaginable form. Agents are not just chatbots or copilots \u2014 they are autonomous software components that can reason about goals, call tools and APIs, collaborate with other agents, and adapt dynamically. Whether it\u2019s a <strong>retrieval agent<\/strong> for research, a <strong>coding agent<\/strong> embedded in a dev workflow, or a <strong>compliance agent<\/strong> ensuring policy enforcement, agents are becoming the next layer of application logic.<\/p>\n<p>Yet despite the excitement, the <strong>path from prototype to production has been fraught with obstacles<\/strong>. Many of the most popular open-source frameworks are fragmented, each with their own APIs and abstractions. Local development rarely maps cleanly to cloud deployments. And most importantly, <strong>enterprise readiness is missing<\/strong>: observability, compliance hooks, security, and long-running durability are table stakes in OSS frameworks.<\/p>\n<p>At Microsoft, we\u2019ve had a front-row seat to this problem. With <strong>Semantic Kernel<\/strong>, we gave developers a stable SDK with connectors into enterprise systems, content moderation, and telemetry. With <strong>AutoGen<\/strong>, pioneered in Microsoft Research, we opened the door to experimental multi-agent orchestration patterns that inspired the community. Both had passionate users \u2014 but each had gaps.<\/p>\n<p>Developers asked us: <em>why can\u2019t we have both \u2014 the innovation of AutoGen and the trust and stability of Semantic Kernel \u2014 in one unified framework?<\/em><\/p>\n<p>That\u2019s exactly why we built the <a href=\"https:\/\/aka.ms\/AgentFramework\"><strong>Microsoft Agent Framework<\/strong><\/a>.<\/p>\n<p>&nbsp;<\/p>\n<p style=\"text-align: center;\"><iframe src=\"\/\/www.youtube.com\/embed\/yOBcPuLLmuY\" width=\"560\" height=\"314\" allowfullscreen=\"allowfullscreen\"><\/iframe><\/p>\n<p>&nbsp;<\/p>\n<h2>Introducing Microsoft Agent Framework<\/h2>\n<p>Microsoft Agent Framework is an <strong>open-source SDK and runtime<\/strong> designed to let developers build, deploy, and manage sophisticated multi-agent systems with ease. It unifies the <strong>enterprise-ready foundations of Semantic Kernel<\/strong> with the <strong>innovative orchestration of AutoGen<\/strong>, so teams no longer have to choose between experimentation and production.<\/p>\n<table style=\"border-collapse: collapse; width: 100%;\">\n<tbody>\n<tr>\n<th style=\"width: 134.365px;\"><\/th>\n<td style=\"width: 530.5px;\"><strong>Semantic Kernel <\/strong><\/td>\n<td style=\"width: 595.531px;\"><strong>AutoGen <\/strong><\/td>\n<td style=\"width: 741.604px;\"><strong>Microsoft Agent Framework<\/strong><\/td>\n<\/tr>\n<tr>\n<td style=\"width: 134.365px;\"><strong>Focus<\/strong><\/td>\n<td style=\"width: 530.5px;\">Stable SDK with enterprise connectors, workflows, and observability<\/td>\n<td style=\"width: 595.531px;\">Experimental multi-agent orchestration from research<\/td>\n<td style=\"width: 741.604px;\">Unified SDK combining innovation + enterprise readiness<\/td>\n<\/tr>\n<tr>\n<td style=\"width: 134.365px;\"><strong>Interop<\/strong><\/td>\n<td style=\"width: 530.5px;\">Plugins, connectors, and support for MCP, A2A, OpenAPI<\/td>\n<td style=\"width: 595.531px;\">Tool integration supported; lacks standardized cross-runtime protocols<\/td>\n<td style=\"width: 741.604px;\">Built-in connectors, MCP + A2A + OpenAPI<\/td>\n<\/tr>\n<tr>\n<td style=\"width: 134.365px;\"><strong>Memory<\/strong><\/td>\n<td style=\"width: 530.5px;\">Multiple vector store connectors and memory store abstraction (e.g. Azure SQL Elasticsearch, MongoDB)<\/td>\n<td style=\"width: 595.531px;\">Support for in-memory \/ buffer history + external vector store memory options (ChromaDB, Mem0, etc)<\/td>\n<td style=\"width: 741.604px;\">Pluggable memory across stores (first-party and third-party), persistent &amp; adaptive memory stored with retrieval, hybrid appraoches<\/td>\n<\/tr>\n<tr>\n<td style=\"width: 134.365px;\"><strong>Orchestration<\/strong><\/td>\n<td style=\"width: 530.5px;\">Deterministic + dynamic orchestration (Agent Framework, Process Framework)<\/td>\n<td style=\"width: 595.531px;\">Dynamic LLM orchestration (debate, reflection, facilitator\/worker, group chat)<\/td>\n<td style=\"width: 741.604px;\">Deterministic + dynamic orchestration (Agent Orchestration, Workflow Orchestration)<\/td>\n<\/tr>\n<tr>\n<td style=\"width: 134.365px;\"><strong>Enterprise readiness<\/strong><\/td>\n<td style=\"width: 530.5px;\">Telemetry, observability, compliance hooks<\/td>\n<td style=\"width: 595.531px;\">Minimal<\/td>\n<td style=\"width: 741.604px;\">Observability, approvals, CI\/CD, long-running durability, hydration<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>With Microsoft Agent Framework, you get:<\/p>\n<ul>\n<li><strong>Open standards &amp; interoperability<\/strong> \u2014 MCP, A2A, and OpenAPI ensure agents are portable and vendor-neutral.<\/li>\n<li><strong>Pipeline for research-to-production<\/strong> \u2014 bleeding-edge orchestration patterns from Microsoft Research are now ready for enterprise use.<\/li>\n<li><strong>Community-driven extensibility<\/strong> \u2014 modular by design, with connectors, pluggable memory, and declarative agent definitions.<\/li>\n<li><strong>Enterprise readiness<\/strong> \u2014 built-in observability, approvals, security, and long-running durability.<\/li>\n<\/ul>\n<p><strong>Microsoft Agent Framework doesn&#8217;t replace Semantic Kernel and AutoGen \u2014 it builds on them.<\/strong> By consolidating their strengths, it gives developers one foundation to move from experimentation to enterprise deployment without compromise. Microsoft Agent Framework supports both <strong>Agent Orchestration<\/strong> (LLM-driven, creative reasoning and decision-making) and <strong>Workflow Orchestration<\/strong> (business-logic driven, deterministic multi-agent workflows). Together, they allow teams to choose the right approach for the problem: flexible collaboration for open-ended tasks, or structured workflows for repeatable enterprise processes.<\/p>\n<p>&nbsp;<\/p>\n<p>Looking ahead, Microsoft Agent Framework further advances integrations across Microsoft\u2019s agent development stack, including the integration with the <strong>Microsoft 365 Agents SDK<\/strong> and a shared runtime with <strong>Azure AI Foundry Agent Service<\/strong>. The Microsoft 365 Agents SDK is the pro-code toolkit that lets developers build full-stack, multi-channel agents and publish them across Microsoft 365 Copilot, Teams, web, and other surfaces, with deep interoperability into Copilot Studio\u2019s low-code connectors and Microsoft 365 Copilot custom engine agents. By converging this SDK with Microsoft Agent Framework\u2014and aligning it with the shared runtime used in Foundry Agent Service\u2014developers will gain <strong>one unified set of abstractions<\/strong> to create, run, scale, and publish agents. This means you can prototype locally, debug with consistent telemetry, and then seamlessly move into scaled hosting with enterprise-grade observability, compliance, and security\u2014all without rewriting your agents\u2014and then publish them into any communication channels of choice where you want to surface your agents.<\/p>\n<p><a href=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-scaled.png\"><img decoding=\"async\" class=\"alignnone wp-image-1354 \" src=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-300x143.png\" alt=\"Screenshot 2025 10 14 214120 image\" width=\"892\" height=\"425\" srcset=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-300x143.png 300w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-1024x489.png 1024w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-768x367.png 768w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-1536x733.png 1536w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/10\/Screenshot-2025-10-14-214120-2048x978.png 2048w\" sizes=\"(max-width: 892px) 100vw, 892px\" \/><\/a><\/p>\n<h2><\/h2>\n<h2>The Four Pillars of Agent Framework<\/h2>\n<ol>\n<li><strong> Open Standards &amp; Interoperability<\/strong><\/li>\n<\/ol>\n<p>Agents don\u2019t exist in isolation \u2014 they need to connect to data, tools, and each other. Microsoft Agent Framework was built with open standards at its core, so developers can choose their integrations and ensure their systems remain portable across frameworks and clouds.<\/p>\n<ul>\n<li><strong>MCP (Model Context Protocol):<\/strong> Agents can dynamically discover and invoke external tools or data servers exposed over MCP. Microsoft Agent Framework makes it easy to connect to a growing ecosystem of MCP-compliant services without custom glue code.<\/li>\n<li><strong>Agent-to-Agent (A2A):<\/strong> Agents can collaborate across runtimes using structured, protocol-driven messaging. A2A support allows developers to create workflows where one agent retrieves data, another analyzes it, and a third validates results \u2014 even if they\u2019re running in different frameworks or environments.<\/li>\n<li><strong>OpenAPI-first design:<\/strong> Any REST API with an OpenAPI specification can be imported as a callable tool instantly. Microsoft Agent Framework handles schema parsing, tool definition, and secure invocation so developers can leverage thousands of enterprise APIs without building wrappers by hand.<\/li>\n<li><strong>Cloud-agnostic runtime:<\/strong> Agents can run in containers, on-premises, or across multiple clouds, making them portable across environments. Developers can spin up a single agent with their preferred SDK (Azure OpenAI, OpenAI, etc.), add tools by wrapping existing methods as AIFunctions, and immediately connect to external APIs.<\/li>\n<\/ul>\n<p>The latest update to the <a href=\"https:\/\/aka.ms\/aitoolkit\"><strong>VS Code AI Toolkit<\/strong> <\/a>brings a streamlined experience for building with the <strong>Microsoft Agent Framework<\/strong>, enabling developers to locally create, run, and visualize multi-agent workflows. These enhancements simplify the inner dev loop, making it easier to build, debug, and iterate on multi-agent systems within the familiar VS Code environment.<\/p>\n<p><a href=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed.jpg\"><img decoding=\"async\" class=\"wp-image-1266 aligncenter\" src=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed-300x161.jpg\" alt=\"VSCode compressed image\" width=\"1057\" height=\"567\" srcset=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed-300x161.jpg 300w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed-1024x550.jpg 1024w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed-768x412.jpg 768w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed-1536x825.jpg 1536w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/VSCode_compressed.jpg 2047w\" sizes=\"(max-width: 1057px) 100vw, 1057px\" \/><\/a><\/p>\n<ol start=\"2\">\n<li><strong> Pipeline for Research<\/strong><\/li>\n<\/ol>\n<p>Microsoft Agent Framework is designed to be the bridge between <strong>research innovation<\/strong> and <strong>enterprise-ready production<\/strong>. Many of the most exciting breakthroughs in multi-agent <a href=\"https:\/\/learn.microsoft.com\/en-us\/azure\/architecture\/ai-ml\/guide\/ai-agent-design-patterns\">orchestration patterns<\/a> come out of Microsoft Research in AutoGen, and the new framework makes those ideas usable in real-world systems without sacrificing durability, governance, or performance.<\/p>\n<p>The framework supports:<\/p>\n<ul>\n<li><strong>Sequential orchestration<\/strong> for step-by-step workflows.<\/li>\n<li><strong>Concurrent orchestration<\/strong> where agents work in parallel.<\/li>\n<li><strong>Group chat orchestration<\/strong> where agents brainstorm collaboratively.<\/li>\n<li><strong>Handoff orchestration<\/strong> where responsibility moves between agents as context evolves.<\/li>\n<li><strong>Magentic <\/strong><strong>orchestration<\/strong> where a manager agent builds and refines a dynamic task ledger, coordinating specialized agents (and sometimes humans) for complex, open-ended problems.<\/li>\n<\/ul>\n<p>To serve both innovators and production-minded developers, Microsoft Agent Framework also provides an <strong>extension package for experimental features<\/strong> \u2014 a clearly labeled incubation channel where advanced users can try out cutting-edge capabilities from Microsoft Research and the open-source community. These features are transparent about their experimental status, while successful innovations graduate naturally into the stable framework.<\/p>\n<p>These patterns \u2014 once prototypes \u2014 now run with durability, auditability, and enterprise controls. It\u2019s the best of research innovation, matured for real-world use.<\/p>\n<p>&nbsp;<\/p>\n<ol start=\"3\">\n<li><strong> Extensible by Design &amp; Community-Driven <\/strong><\/li>\n<\/ol>\n<p>Microsoft Agent Framework is <strong>100% open source<\/strong> and designed to grow with the community. Its modular design makes it easy to extend, customize, and contribute.<\/p>\n<ul>\n<li><strong>Connectors to enterprise systems:<\/strong> Agent Framework inherits a broad set of built-in connectors (Azure AI Foundry, Microsoft Graph, Microsoft Fabric, SharePoint, Oracle, Amazon Bedrock, MongoDB, and a various SaaS system through Azure Logic Apps) so agents can work with enterprise data from day one.<\/li>\n<li><strong>Pluggable memory modules:<\/strong> Developers can choose <a href=\"https:\/\/aka.ms\/Redis\/AFBlog\">Redis<\/a>, Pinecone, Qdrant, Weaviate, Elasticsearch, Postgres, or their own store for conversational memory. Agent Framework provides the abstraction; you decide the backend.<\/li>\n<li><strong>Declarative agents<\/strong><strong>:<\/strong> YAML or JSON definitions allow developers to specify prompts, roles, and tools declaratively. These files can be version-controlled, templatized, and shared across teams.<\/li>\n<li><strong>Community innovation:<\/strong> Agent Framework is designed to absorb community-driven orchestration strategies, new connectors, and best practices.<\/li>\n<\/ul>\n<p>This means Microsoft Agent Framework is not a fixed product \u2014 it\u2019s a <strong>living ecosystem<\/strong>, continuously shaped by contributions from Microsoft Research and the global OSS community.<\/p>\n<p>&nbsp;<\/p>\n<ol start=\"4\">\n<li><strong> Ready for Production<\/strong><\/li>\n<\/ol>\n<p>Microsoft Agent Framework isn\u2019t just for experimentation \u2014 it was built for enterprise-grade deployment from the very beginning. It delivers the end-to-end tooling and runtime features needed to confidently move from prototype to scale, while integrating deeply with the Azure AI Foundry ecosystem.<\/p>\n<ul>\n<li><strong>Observability:<\/strong>\u00a0<a href=\"https:\/\/aka.ms\/MultiAgentTracingBlog\">OpenTelemetry<\/a> can instrument and visualize every agent action, tool invocation, and orchestration step, making it easy to trace reasoning flows and monitor performance through Azure AI Foundry dashboards.<\/li>\n<li><strong>Secure Cloud Hosting:<\/strong> Agents will run natively on Azure AI Foundry with enterprise controls like virtual network integration, role-based access, private data handling, and built-in content safety.<\/li>\n<li><strong>Security &amp; compliance:<\/strong> Azure AI Content Safety integration, Entra ID authentication, and structured logging mean Agent Framework agents can run in regulated industries.<\/li>\n<li><strong>Long-running durability:<\/strong> Agent threads and workflows can pause, resume, and recover from interruptions, with retry and error-handling logic ensuring long-running processes remain reliable at scale.<\/li>\n<li><strong>Human in the loop:<\/strong> For scenarios that require governance, tools can be marked as requiring human approval. Agent Framework automatically emits a pending approval request that can be routed to a UI or queue, then continues or denies execution accordingly. This works across local tools or remote service calls, ensuring sensitive operations remain under control.<\/li>\n<li><strong>CI\/CD integration:<\/strong> The framework integrates directly into GitHub Actions and Azure DevOps pipelines, with telemetry flowing into Azure Monitor and Application Insights for enterprise-grade deployment and root-cause analysis.<\/li>\n<\/ul>\n<p>With these capabilities, Microsoft Agent Framework makes it seamless to <strong>prototype locally, debug with rich telemetry, and then scale securely into production<\/strong> with the enterprise readiness that modern AI systems demand.<\/p>\n<h2><\/h2>\n<h2>Customer Momentum<\/h2>\n<p>Enterprises across industries are already testing Microsoft Agent Framework in real-world scenarios:<\/p>\n<p><a href=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers.png\"><img decoding=\"async\" class=\"wp-image-1267 aligncenter\" src=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers-300x114.png\" alt=\"AgentFramework customers image\" width=\"942\" height=\"358\" srcset=\"https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers-300x114.png 300w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers-1024x388.png 1024w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers-768x291.png 768w, https:\/\/devblogs.microsoft.com\/foundry\/wp-content\/uploads\/sites\/89\/2025\/09\/AgentFramework-customers.png 1210w\" sizes=\"(max-width: 942px) 100vw, 942px\" \/><\/a><\/p>\n<ul>\n<li><a href=\"https:\/\/aka.ms\/kpmgagentframework\"><strong>KPMG<\/strong><\/a> is building <em>Clara AI<\/em>, a multi-agent system that automates audit testing and documentation. <em>\u201cFoundry Agent Service and Microsoft Agent Framework connect our agents to data and each other, and the governance and observability in Azure AI Foundry provide what KPMG firms need to be successful in a regulated industry\u201d<\/em> &#8211; Sebastian St\u00f6ckle, Global Head of Audit Innovation and AI, KPMG International<\/li>\n<li><strong>Commerzbank <\/strong>is piloting Microsoft Agent Framework to power avatar-driven customer support, enabling more natural, accessible, and compliant customer interactions. <em>\u201cThe new Microsoft Agent Framework simplifies coding, reduces efforts and fully supports MCP for agentic solutions. We are really looking forward to the productive usage of container-based Azure AI Foundry agents, which\u00a0significantly reduces workload in IT operations\u201d<\/em> \u2013 Gerald Ertl, Managing Director\/Head of Digital Banking Solutions, Commerzbank AG<\/li>\n<li><strong>BMW:<\/strong> BMW is using Microsoft Agent Framework and Foundry Agent Service to orchestrate multi-agent systems that analyze terabytes of vehicle telemetry in near real time, enabling engineers to accelerate design cycles and spot issues earlier in testing. <em>\u201cDurability and observability are key for our operations. With multi-agent systems powered by Microsoft Agent Framework and Foundry Agent Service, engineers don\u2019t just access data \u2014 they get insights they can act on immediately, cutting analysis from days to minutes.\u201d &#8211; <\/em>Christof Gebhart, Manager, Advanced Vehicle Measurement Technology, BMW<\/li>\n<li><strong>Fujitsu<\/strong> is embedding Microsoft Agent Framework into its integration services, enabling customers to safely adopt advanced orchestration strategies such as group chat and debate. <em>\u201cI believe it enables us to build multi-agent systems that emphasize coexistence between humans and AI and can truly accelerate our AI transformation\u201d \u2013 <\/em>Hirotaka Ito, Lead engineer of AI, Corporate Digital Unit,\u00a0Fujitsu.<\/li>\n<li><strong>Citrix: <\/strong>Citrix is exploring how they can use agentic AI within VDI environments to improve enterprise productivity and efficiency<strong><span data-olk-copy-source=\"MessageBody\">.<\/span><\/strong>\u00a0<em>\u201cWe are excited about the Microsoft Agent Framework, which brings a modern, developer-first approach to building single- and multi-agent workflows. With support of key APIs and languages, and native adoption of emerging protocols for tool calling and observability, it enables intuitive development of agents on Azure AI Foundry, without compromising developer control. We are eager to leverage the framework to deliver enterprise-scale, production-ready AI solutions to our customers.\u201d \u2014 <\/em>George Tsolis, Distinguished Engineer, Citrix<\/li>\n<li><strong>Fractal:<\/strong> Fractal\u2019s <strong>Cogentiq<\/strong> is an agentic AI platform that uses Microsoft Agent Framework to orchestrate and scale enterprise AI agents and workflows across industries.<em> \u201cCogentiq leverages Microsoft Agent Framework to orchestrate and scale AI agents and workflows. Microsoft Agent framework&#8217;s ease of use, flexible agent development and deployment options and support for building complex multi-agentic workflows enables us to rapidly build, deploy, and manage multi-agent solutions across industries and functions. The Agent Framework allows both technical and business teams to innovate quickly, integrate with enterprise systems, and deliver value at scale through production-ready tools and access to industry-leading AI models. It takes care of the heavy lifting around model access, deployment, scaling, security, networking helping Fractal focus on solving our client&#8217;s industry and function specific business problems.\u201d \u2014 <\/em>Himanshu Nautiyal, Chief Product Officer, Fractal<\/li>\n<li><strong>TCS:<\/strong> Tata Consultancy Services is actively building a multi-agent practice on the Microsoft Agent Framework, with several initiatives underway that showcase their strategic investment and technical depth including agentic solutions for finance, IT operations, and retail.<em> &#8220;Adopting Microsoft Agent Framework is not just a technological advancement, but a bold step towards reimagining industry value chains. By harnessing Agentic AI and Frontier models, we enable our teams to build flexible, scalable, enterprise-grade solutions that transform workflows and deliver value across platforms. True leadership is about empowering innovation, embracing change, and fostering an environment where agility and collaboration drive excellence.&#8221;\u00a0 &#8211; Girish Phadke, Head, Microsoft Azure Practice, TCS<\/em><\/li>\n<li><strong>Sitecore:<\/strong> Sitecore is building a solution to help marketers interact more seamlessly with the Sitecore platform by automating tasks across content supply chain, from creating and managing web experiences to digital assets, using intelligent agents. <span style=\"font-style: normal !msorm;\"><em>\u201cBy partnering with Microsoft to leverage its new<\/em><\/span><em> Microsoft<span style=\"font-style: normal !msorm;\"> Agent Framework, Sitecore can bring together the best of both worlds: the flexibility to power fully non-deterministic agentic orchestrations and the reliability to run more deterministic, repeatable agents. At the same time, we benefit from Microsoft\u2019s enterprise-grade observability and telemetry, ensuring that these orchestrations are not only powerful but also secure, measurable, and production-ready.\u201d<\/span><\/em> &#8211; Mo Cherif, VP of AI, Sitecore.<\/li>\n<li><strong>NTT<\/strong><strong> DATA:<\/strong> Agentic AI value includes a complete <a href=\"https:\/\/nam06.safelinks.protection.outlook.com\/?url=https%3A%2F%2Furldefense.com%2Fv3%2F__https%3A%2Fservices.global.ntt%2Fen-US%2Fnewsroom%2Fntt-data-unveils-smart-ai-agent-ecosystem__%3B!!AKgNyAfk550h-spHnEo!diUrQDvWM54M4vZt1ndl0xHwQiU7NJ0TZF9FQmsGmaB0Q6CNxePS6nvTNkg2IQ9etx12UFbOg3ophlUKNkpxnfdPLK9MTr_aNC0%24&amp;data=05%7C02%7Cjmachado%40microsoft.com%7C20e6e0dbfc6043e6467308ddfc75de40%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C638944305574627620%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&amp;sdata=XEfA79bN%2BZ6Bzzbvy0tBICeYMjhr0QbqXadHyQrR3aw%3D&amp;reserved=0\">ecosystem<\/a> of solutions, services, and partners. NTT DATA is adopting the Microsoft Agent Framework in alignment with efforts to standardize its R&amp;D approach for multi-agent management, enabling the company to deploy, manage, and optimize AI solutions across industries. This will help accelerate deployments, support complex process workflows that can be customized and replicated and make it easier to connect and orchestrate sophisticated models on behalf of clients. \u201cBy adopting the Microsoft Agent Framework, NTT DATA is not only further standardizing how we develop and manage multi-agent systems, but also accelerating how our clients realize value from AI. This initiative allows us to deliver faster, more scalable, and more governed AI solutions, while staying closely in step with Microsoft\u2019s engineering roadmap.\u201d- Charlie Doubek, Global VP, Agentic AI Services Leader, Cloud and Security, NTT DATA<\/li>\n<li><strong>MTech Systems:<\/strong> MTech Systems will use the new Agent Framework to orchestrate transactional data anomaly sweeps, human-in-the-loop approvals, and automated fixes \u2013 agent patterns that previously required extensive glue code. <em>\u201cThe framework gives us a batteries-included developer experience and makes agent workflows far easier to build and run. Features like checkpointing and declarative YAML workflows will save us time and let us scale changes across hundreds of customer applications without redeploys<\/em><em>\u201d \u2013 Barry Schulz, CTO, MTech Systems<\/em><\/li>\n<li><strong>TeamViewer<\/strong>: TeamViewer is embedding agentic AI into its IT support stack so that remote support agents can get real-time diagnostics, automated summarization, and contextual recommendations during sessions. <em>\u201cThe framework strikes the right balance between technical depth and usability. Its intuitive design and modular structure make it easy for our teams to adopt quickly, while providing the scalability and flexibility we need for complex projects. That combination allows us to deliver value faster today and positions us well to take advantage of the enhancements still to come.\u201d<\/em>&#8211; Mei Dent, Chief Product and Technology Officer TeamViewer<\/li>\n<li><strong>Weights &amp; Biases<\/strong>: Weights &amp; Biases is collaborating with Microsoft to ensure developers can seamlessly train, track, and operationalize AI agents at scale. <em>&#8220;The new Microsoft Agentic Framework makes building production-ready agents dramatically easier. With flexible orchestration, checkpointing to save time and compute, and built-in human-in-the-loop support, it tackles the real challenges teams face when moving from prototype to production &#8220;<\/em> &#8211; Phil Gurbacki, VP of Product, Weights &amp; Biases<\/li>\n<li><strong>Elastic<\/strong>:\u00a0Elasticsearch supports a native connector to Microsoft Agent Framework, enabling developers to seamlessly integrate enterprise data into intelligent agents and workflows. <em>\u201cElasticsearch is the context engineering platform and vector database of choice for organizations to store and search their most valuable operational and business data. With the new Microsoft Agent Framework connector, developers can now bring the most relevant organizational context directly into intelligent agents and multi-agent workflows. This makes it easier than ever to build production-ready AI solutions that combine the reasoning power of agents with the speed and scale of Elasticsearch.\u201d<\/em> \u2014 Steve Kearns, General Manager Search Solutions, Elastic<\/li>\n<\/ul>\n<p>These early stories highlight the dual promise of Microsoft Agent Framework: <strong>innovative enough to inspire new approaches, stable enough to deploy in production.<\/strong><\/p>\n<p>&nbsp;<\/p>\n<h2>Path to Microsoft Agent Framework<\/h2>\n<p>Many customers are already using <strong>Semantic Kernel<\/strong> or <strong>AutoGen<\/strong> in production today. Both projects will remain supported but most investment is now focused on Microsoft Agent Framework. Developers using <strong>Semantic Kernel<\/strong> or <strong>AutoGen<\/strong> will find the transition straightforward:<\/p>\n<ul>\n<li>For <strong>Semantic Kernel<\/strong> users:\n<ul>\n<li>Migration is straightforward: replace Kernel and plugin patterns with the <strong>Agent<\/strong> and <strong>Tool<\/strong> abstractions.<\/li>\n<li>.NET developers move from Microsoft.SemanticKernel.* to the new Microsoft.Extensions.AI.* namespaces, with agents created directly from providers instead of requiring Kernel coupling.<\/li>\n<li>Python developers can install the full package (pip install agent-framework) or just the components they need (e.g., agent-framework-azure-ai, agent-framework-redis).<\/li>\n<li>Agents now manage threads natively, simplify invocation with RunAsync \/ RunStreamingAsync, and register tools inline without attributes or plugin wrappers.<\/li>\n<li>Existing vector store integrations (Azure AI Search, Postgres, Cosmos DB, Redis, Elasticsearch, etc.) continue to work through connectors.<\/li>\n<li>Plugins like Bing, Google, OpenAPI, and Microsoft Graph port directly as tools, often exposed via MCP or OpenAPI.<\/li>\n<li>The net result: <strong>less boilerplate, simplified memory management, and alignment with open standards.<\/strong><\/li>\n<\/ul>\n<\/li>\n<li>For <strong>AutoGen<\/strong> users:\n<ul>\n<li>AutoGen pioneered many orchestration patterns (GroupChat, GraphFlow, event-driven runtimes), which are now unified in Agent Framework under the <strong>Workflow<\/strong> abstraction.<\/li>\n<li>The AssistantAgent maps directly to the new ChatAgent, which is multi-turn by default and continues tool invocation until a result is ready.<\/li>\n<li>FunctionTool wrappers migrate to the @ai_function decorator, with automatic schema inference and support for hosted tools like code interpreter or web search.<\/li>\n<li>Messaging is simplified: multiple message classes are replaced with a unified ChatMessage type, with explicit roles (USER, ASSISTANT, TOOL, SYSTEM).<\/li>\n<li>Orchestration shifts from event-driven models to a typed, graph-based Workflow API that supports checkpointing, pause\/resume, and human-in-the-loop flows.<\/li>\n<li>Observability is richer and simpler, with OpenTelemetry support out of the box.<\/li>\n<li>Most single-agent migrations require only light refactoring; multi-agent migrations benefit from the new Workflow model with stronger composability and durability.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p>This continuity means developers can preserve their existing investments while unlocking new capabilities. Microsoft Agent Framework is not a replacement for what came before \u2014 it is the natural evolution that unites innovation and stability. For more information about migration, see the <a href=\"https:\/\/learn.microsoft.com\/en-us\/agent-framework\/overview\/agent-framework-overview\">documentation<\/a>.<\/p>\n<p>&nbsp;<\/p>\n<h2>Get Started with Microsoft Agent Framework Today<\/h2>\n<p>Agents are fast becoming the next layer of application logic \u2014 reasoning about goals, calling tools, collaborating with each other, and adapting dynamically. With Microsoft Agent Framework, developers now have a single, open-source foundation that carries the best of research innovation into production with the durability, observability, and enterprise readiness required to scale.<\/p>\n<p>This is the natural evolution of the journey that began with Semantic Kernel and AutoGen \u2014 and it\u2019s only the beginning. By building in the open and co-creating with the developer community, Microsoft Agent Framework will continue to evolve as the foundation for next-generation multi-agent systems.<\/p>\n<p>&nbsp;<\/p>\n<ul>\n<li style=\"list-style-type: none;\">\n<ul>\n<li><strong>Download the SDK:<\/strong> <a href=\"https:\/\/aka.ms\/AgentFramework\">aka.ms\/AgentFramework<\/a><\/li>\n<li><strong>Explore documentation for more details:<\/strong> <a href=\"https:\/\/aka.ms\/AgentFramework\/Docs\">https:\/\/aka.ms\/AgentFramework\/Docs<\/a><\/li>\n<li><strong>See it in action:<\/strong> Watch demos on <a href=\"https:\/\/aka.ms\/AgentFramework\/AIShow\">AI Show<\/a> and <a href=\"https:\/\/aka.ms\/AgentFramework\/OpenAtMicrosoft\">Open at Microsoft <\/a><\/li>\n<li><strong>Learn step by step<\/strong>: <a href=\"https:\/\/learn.microsoft.com\/en-us\/training\/paths\/develop-ai-agents-on-azure\/\">Microsoft Learn modules for Agent Framework<\/a> and <a href=\"https:\/\/github.com\/microsoft\/ai-agents-for-beginners\">AI Agents for Beginners<\/a><\/li>\n<li><strong>Join the <\/strong><a href=\"https:\/\/aka.ms\/foundry\/discord\"><strong>Azure AI Foundry Discord<\/strong><\/a> to connect with developers and product groups, sharpen your AI skills, and stay inspired through real-time community. Join us for an AMA Tuesday 7<sup>th<\/sup> October 9am PST: <a href=\"https:\/\/aka.ms\/foundry\/discord\">https:\/\/aka.ms\/foundry\/discord<\/a><\/li>\n<\/ul>\n<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Why agents need a new foundation Over the last year, developers have been experimenting with AI agents in every imaginable form. Agents are not just chatbots or copilots \u2014 they are autonomous software components that can reason about goals, call tools and APIs, collaborate with other agents, and adapt dynamically. Whether it\u2019s a retrieval agent [&hellip;]<\/p>\n","protected":false},"author":190969,"featured_media":1285,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[50,49,51,1],"tags":[25,3,5,10,12,2,23],"class_list":["post-1260","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-a2a","category-aiagent","category-mcp","category-microsoft-foundry","tag-agents","tag-ai-development","tag-ai-tools","tag-ai-agents","tag-azure-openai","tag-microsoft-foundry","tag-semantic-kernel"],"acf":[],"blog_post_summary":"<p>Why agents need a new foundation Over the last year, developers have been experimenting with AI agents in every imaginable form. Agents are not just chatbots or copilots \u2014 they are autonomous software components that can reason about goals, call tools and APIs, collaborate with other agents, and adapt dynamically. Whether it\u2019s a retrieval agent [&hellip;]<\/p>\n","_links":{"self":[{"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/posts\/1260","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/users\/190969"}],"replies":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/comments?post=1260"}],"version-history":[{"count":0,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/posts\/1260\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/media\/1285"}],"wp:attachment":[{"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/media?parent=1260"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/categories?post=1260"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/foundry\/wp-json\/wp\/v2\/tags?post=1260"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}