TypeSafe AI's System One model for fast, structured decisions such as choices, scores, and boolean judgments.
An open protocol for connecting AI applications and agents to tools, data, and resources.
Designing products, APIs, documentation, permissions, and workflows for AI agents as first-class users.
Identity and authorization primitives that let an agent act as a distinct principal with scoped permissions.
Persistent state and memory that let agents retain context, preferences, and progress across interactions.
Payment and transaction infrastructure designed for software agents that can discover, buy, and pay autonomously.
A protocol for interoperability and communication between independent AI agents.
Infrastructure designed around autonomous software agents rather than only human-operated applications.
Vercel's TypeScript AI SDK release focused on production agents, harnesses, approvals, sandboxes, MCP Apps, voice, and video.
A Cloudflare SDK and runtime model for building stateful agents with persistence, realtime communication, and tools.
An extension pattern that lets MCP-connected tools expose interactive application UI alongside model-visible capabilities.
Constrained model outputs that conform to a defined schema, making AI decisions easier to integrate into software.