Built my first SaaS as a frontend dev with no backend experience using a similar approach. The key shift was treating Claude Code as a senior developer who needs clear specs, not a magic box. The more precise the context and requirements, the better the output. Vague prompts produce vague code.

Rule 5 is the one that took me longest to internalize. Coming from frontend development into building a full product with a real database, I kept reaching for complex query logic when the real fix was just restructuring the data. Once the schema was right the queries became obvious. Brooks was right 50 years ago and it's still true.

The interesting angle here is what this means for passes and credentials stored in Apple Wallet. If device compromise is this accessible, the assumption that Wallet passes are isolated from the rest of the device needs more scrutiny. Apple's security model relies heavily on the secure enclave but a tool like this changes the threat surface significantly.

This is always the threat with walled garden style security. When you couple applications so tightly in an intrinsic trust network, on the basis that no external attacker can gain access, then the internal security is neglected and it only takes the weakest link.

Nice idea, pricing comparison across providers is genuinely painful to do manually right now. One feature that would make this more useful for me — a cost estimator based on expected token volume or GPU hours, not just raw pricing. Context switching between providers to calculate actual monthly cost is where the friction is.

Switched to Claude Code a while ago and barely open a traditional IDE anymore. Features like "Code With Me" made sense before AI coding tools existed. Hard to see how JetBrains competes on collaboration features when the whole development workflow is shifting.

Using MCP daily as a solo founder with Claude Code. The "consistent rails" point resonates. The value isn't just tool calling, it's that the agent knows how to behave within a defined boundary. The security posture argument is underrated too. Giving a CLI unrestricted box access vs a hosted MCP server with scoped permissions is a completely different risk profile.

The zero-knowledge proof angle is interesting but the real barrier is implementation

most platforms won't voluntarily adopt privacy-preserving verification when the surveillance version gives them more data. Regulation would need to mandate the privacy-preserving approach specifically, not just "verify age somehow.

This is the Gettysburg Address run through a corporate buzzword generator. Funny how Lincoln managed to write one of the most memorable speeches in history in 272 words, while modern communication trends push us toward adding more words to say less.