I'm more and more hesitant to use third party harnesses.
It's so easy to take opencode or pi and build my own for exactly what I want that I'm seeing projects like this to be noisy and less valuable over time.
I wish this wasn't the case. I love software, but it's becoming increasingly easy to just build my own.
I'm sick of AI slop, and decided to do something about it.
Even with Fable I'm still seeing wrong but plausible answers, which are the worst kind imo.
AGENTS.md didn't quite give me the right boost either, so here's my flow:
1. Asking 2 new sessions to review what I built in my other session along with guides/standards*
2. Having a third session judge the first two. It dedupes, and does a good job of only surfacing the important stuff
3. Ask it to make edits, etc.
Some missing parts that I've been finding improve review are:
- current/past session context
- indexed codebase
- PRs/git history
- research/guides/3rd party context*
I had been running this on my own for a long time, and just started to turn it into a product.
It's free right now, would love any feedback (roast me). Looking forward to catching some bugs for you guys.
I owned two MD players and used it for songwriting, along with a small directional Sony mic with a mini jack.
This was the peak of non-DAW editing. The quality was incredible, and if SSDs weren’t around the corner, I think this tech would have had a chance.
I'm worried about the environmental impacts of this, but from everything I've seen society values model output more. Curious to watch this over the rest of the decade.
Express is extremely mature as a backend, although you may not like the middleware patterns compared to Flask, for example. I've used Flask for years, and find these similar, with some differences.
Next.js + Vercel has strong adoption. This probably comes the closest to checking all of your requirements, although I wouldn't suggest using this for all use cases.
IMO the Node backend framework ecosystem is more mature than Python (no idea about .NET) since Python really is positioned better for data science/ML, etc.
Express itself might be mature but it's a fraction of what you need to make a full application.
The JS backend ecosystem is definitely not mature. Eg: ORMs are still being figured out. People are migrating en masse to Drizzle which hasn't even reached 1.0. The best (fastest, better features, etc) PG driver for JS is Postgres.js which is maintained primarily by a single guy. Etc.
Next is a bad option IMO. They keep experimenting while having users on board and introducing breaking changes. The performance for SSR is terrible. Hosting it outside of Vercel is a road full pain. Etc.
I've never used Python but Django (or Rails, or .NET, etc) are more mature than most of the NPM ecosystem.
It sounds like you're looking for a Django/Rails type framework for JS-- you're not really going to find this. Javascript isn't really the ecosystem that's going to give you something opinionated, although it'd be a nice addition.
I'm not sure you're giving JS it's due. Prisma is an excellent ORM-- it makes Sqlalchemy feel clunky. I'm not sure about your comment on SSR performance either, although I do agree that Vercel tends to be a lot of pixie dust on top of EC2.
I didn't dig into this deeply, but I miss this aesthetic more than I'd like to admit. I'm sure it reflects on simpler times, but there's some comfort in seeing projects like this for me.
It's great to see more and more adoption for MCP. I'm not sure it's the most bulletproof protocol, but it feels like it's in a strong lead, especially with OpenAI support.
I've been using Codex for the last 24 hours, and background mode boosts your output. You can have Codex work on n+ features async. I had it building a database model alongside frontend authentication, and did both pretty well.
I wish this wasn't the case. I love software, but it's becoming increasingly easy to just build my own.