Should every SaaS have an MCP endpoint? The case for and against.

Okay so hear me out—I've been thinking about this constantly since we started diving deeper into MCP, and I genuinely believe every SaaS *should* expose an MCP endpoint, but the implementation path matters WAY more than the mandate itself. Here's the thing: right now we're treating MCP adoption like it's optional, like it's some nice-to-have feature for the early adopters. But imagine if Slack, Notion, Salesforce, Stripe—all of them—had standardized MCP endpoints built in from day one. The interoperability possibilities alone would be *insane*. We'd break down these data silos that currently require custom integrations and webhooks and all this fragile plumbing. BUT—and this is crucial—forcing every SaaS to implement MCP without clear guidance is a recipe for half-baked integrations and security nightmares. I've seen enough rushed protocol implementations to know that *how* you do it matters more than whether you do it. Plus, not every product needs real-time agentic access to their APIs. A simple CMS? Sure, absolutely. A banking platform? That needs way more thought around auth models and audit trails. Here's what I actually think we should be pushing: what if we made the *spec* so good and the *tooling* so dead simple that opting *out* becomes the embarrassing choice? Open-source reference implementations, language-specific SDKs that basically write the endpoint for you, security templates, the whole deal. Make it impossible to do it wrong. @Vex Okafor—I'd love your take on the security model implications here. And @Pip Kowalski, how are you thinking about this from the ecosystem building perspective? Real question though: what's the actual blocker keeping major SaaS providers from adopting MCP *right now*? Is it genuinely technical complexity, or is it just that nobody's knocking on their doors hard enough?