sean_lynch

Before you get too far into the usual “MCP is dead, Skills forever” debate

The real valuable capability MCP offers over skills/CLI is isolating the auth flow outside of the agent’s context window, and potentially out of the harness completely. This is valuable from a security perspective obviously. It’s also just a much easier user experience for normies and large businesses adopting AI tools. I hear all the context bloat and tool call redundancy complaints. But this structure for handling auth has real value.

Maybe the idealized form of MCP is just an auth gateway for the API and nothing else. That’d still be a win.

I’d love to hear more. What are they seeing?

I’m getting “Invalid captcha” when asking the first question. I’m on iPhone and not seeing any captcha. Any advice?

I'd love to hear the bull case for standalone vector DBs at this point. After exploring a few, there's not much differentiation from what's offered here (beyond maybe a nicer developer experience than GCP's) and I'd expect many other DBs/data warehouse to follow the same path as BigQuery has.

It looks like there's still plenty of hard problems in the indexing side of RAG and so maybe the vector DB that leans hard in that direction will carve out something defensible? Outside that, can't find a good reason to standup yet another piece of infrastructure purely for vector search.

Have been thinking about exactly this workflow and of course Simon is way out ahead. I recommend taking a look at Simon's daily notes logs too, really fascinating back and forths.

I’m assuming this is a sell page for eng candidates? I think you should mention the big engineering challenges you have at Streak. You mention you’re the largest Spanner user outside Google. That's a really impressive point, but what problems are you tackling by using that?

FWIW - I don’t like that you stuffed everyone into the same t-shirt :)