
MCP
MCP (Model Context Protocol) is the open standard that lets AI agents reach your software through a typed tool interface. These notes cover what it is, when building a server is worth it, and the patterns that hold up in production.
Start here


Vetting third-party MCP servers: 7 checks before production
The official MCP Registry listed 53,756 active servers on 2 November 2026 and scans none of them. Seven checks to run before one reaches production.
Figma MCP design-to-code handoff: 6 steps and its limits
The Figma MCP server gives coding agents design context and a code starting point. Six steps to a clean handoff, and the four gaps it still leaves.
WebMCP in 2026: the browser API that lets agents use your SaaS UI
WebMCP lets a web page register tools that browser AI agents call directly. Chrome 149 runs it as an origin trial. What it changes for SaaS interfaces.
MCP governance in 2026: what the Linux Foundation handoff changed
Anthropic donated MCP to the Linux Foundation in December 2025. Who decides the spec now, what the 2026-07-28 release removed, and what to check.
Harden an MCP server for enterprise: SSO, audit, gateway
A 2026 scan found 33% of MCP servers had critical vulnerabilities. Here is how to add OAuth 2.1 SSO, audit trails, and a gateway to yours.
Short-lived MCP server URLs: the TTL pattern for AI agents in 2026
1,467 MCP servers are exposed on the public internet with no auth, per Trend Micro. Short-lived URLs with a TTL of minutes, not days, close the hole.
Managed vs self-hosted MCP server: 6 axes that decide it
Time to first tool, auth, security posture, cost at scale, data residency, protocol velocity. The six axes that decide whether you build an MCP server or buy.