Blog

Why some user interactions cannot safely happen inside the MCP client, and how URL-mode elicitation helps.

The naive approach to RAG authorization creates a scaling nightmare. Here's the pattern that actually works.

A practical guide to onboarding organizations, admins, and users at scale.

Fireworks.ai is betting that inference is the real AI runtime. A look at its PyTorch roots, serving stack, and compound model strategy.

WorkOS Pipes, AuthKit for Platforms, Multiple Roles for SSO and SCIM, & more

The hidden stack that lives between a model that works and a product that scales.

The spec evolution that made zero-config MCP connections possible—and closed a category of security holes

How enterprise software teams can ship faster without breaking trust

Understand how OAuth clients authenticate and when you should use client secrets vs private key JWT.

Build a CIMD-based confidential MCP client in Python using Authorization Code + PKCE.

Anthropic, OpenAI, Block, and the Linux Foundation discuss governing MCP together as the new Agentic AI Foundation launches with 50 companies on day one.

Michael Grinich traces MCP's evolution from local file system interface to industry standard, announces the Agentic AI Foundation, and walks through the latest spec updates.

Paul Irish shows how Chrome DevTools' MCP integration lets AI agents parse 15M-line performance traces and debug browser sessions programmatically.

Den Delimarsky demonstrates MCP's new auth flow—Protected Resource Metadata replaces Dynamic Client Registration for zero-config authentication.

Reilly Wood on why structured queries beat freeform commands at scale