An agent or workflow is only useful when people and other systems can reach it through the interfaces and channels they already use. That might be an OpenAI Responses client, Telegram, another agent using A2A, or an MCP client. As a builder of agents and workflow, you need control over which channels you expose it to and how it behaves. To help, ...

Prompt injection is the #1 risk on the OWASP LLM Top 10, and most agents in production today defend against it with one of two heuristics: a defensive system prompt, or a hand-rolled allowlist. Neither is deterministic. Both fail silently the day someone slips a line into an issue body, an email, or a tool result. With FIDES (Flow Integri...

Modern AI agents often aren't bottlenecked by model quality, they are bottlenecked by orchestration overhead. When an agent chains together many small tool calls, each step typically requires a new model turn, driving up latency and token usage. With CodeAct support in Agent Framework, agents can collapse those multi-step plans into a sing...

Handling Long-Running Operations with Background Responses AI agents powered by reasoning models can take minutes to work through complex problems — deep research, multi-step analysis, lengthy content generation. In a traditional request-response pattern, that means your client sits idle waiting for a connection that may time out, or worse, fails ...

Agent harness is the layer where model reasoning connects to real execution: shell and filesystem access, approval flows, and context management across long-running sessions. With Agent Framework, these patterns can now be built consistently in both Python and .NET. In this post, we’ll look at three practical building blocks for production agent...

We're excited to announce a significant update to Semantic Kernel Python's vector store implementation. Version 1.34 brings a complete overhaul that makes working with vector data simpler, more intuitive, and more powerful. This update consolidates the API, improves developer experience, and adds new capabilities that streamline AI development work...

We are excited to announce that Semantic Kernel (SK) now has first-class support for the Model Context Protocol (MCP) — a standard created by Anthropic to enable models, tools, and agents to share context and capabilities seamlessly. With this release, SK can act as both an MCP host (client) and an MCP server, and you can leverage these capabili...

Announcing New Vector Stores: Faiss, SQL Server, and Pinecone We are thrilled to announce the availability of three new Vector Stores and Vector Store Record Collections: Faiss, SQL Server, and Pinecone. These new connectors will enable you to store and retrieve vector data efficiently, making it easier to work with your own data and data models...

Introducing Realtime Agents in Semantic Kernel for Python With release 1.23.0 of the Python version of Semantic Kernel we are introducing a new set of clients for interacting with the realtime multi-modal API's of OpenAI and Azure OpenAI. They provide a abstracted approach to connecting to those services, adding your tools and running apps that le...

We are thrilled to announce the General Availability (GA) of our Memory packages for .NET, Java, and Python! The Semantic Kernel team and our partners have been working hard to allow you to quickly connect to vector stores and make it easy for you to do embeddings and data retrieval between stores. Transforming Data Management with Vector Stores...