Astra has a serious problem of having zero respect for my github actions limits. One shot project, I didn't even see the result yet, and each run already takes 1h30m
I'll write a proper post mortem from this experiment later, but in summary.
- Code looks bad, `unsafe` everywhere, likely an attempt to stay 1 to 1 with a C code base. Which is a fair.
- It found 6 real bugs on QuickJs, which I'll upstream later
- It implemented everything, including Regex and Unicode.
- It is SO F* COOL...
As the @twominutepapers guy says, what a time to be a live.
Note: I hate @ChatGPT cybersecurity checks with all my heart
Twitter is a hurtful place even when you come here just to share the silly things you vibed on your free time.
1 1/2 day later:
✅ Compression algorithms
✅ HTTP/2
⚠️ WASI 1/2, needs some cleanup
😱 QuickJs rewrite. Ugliest code and backlog grows larger every time I look, but let's keep pushing..
What happens if ask Astra on Extra High when your resets are about to expire:
- Full rewrite of QuickJs in Silkla.ng
- LZMA, LZMA2, and XZ decoding
- Raw DEFLATE, zlib, and gzip encoding.
- full HTTP/2 support
- WASI 1/2 support as a target?
Let's find out!
What happens if ask Astra on Extra High when your resets are about to expire:
- Full rewrite of QuickJs in Silkla.ng
- LZMA, LZMA2, and XZ decoding
- Raw DEFLATE, zlib, and gzip encoding.
- full HTTP/2 support
- WASI 1/2 support as a target?
Let's find out!
writing a compiler is a sequence of :
"Oooooh, thats why they did [thing that I hate in another language] this way"
How should the Silk book be called (by book I mean just a guide in the docs)? One shot covers by Astra 😆
Wow… Delta (@zeddotdev) is pretty cool. I try a new harness pretty much every day, and it’s nice to see someone trying something different.
My biggest pet peeve with them is that AI already puts some distance between devs and their code, but most harnesses seem to go out of their way to keep us as uninvolved as possible. Delta seems to get that.
J�lia retweeted
The latest @foldkit Blocks + Effect Layers preview is ready. I’ll keep refining it as feedback comes in.
What’s new:
- Simpler root setup: compose your application from Blocks, then pass the root ApplicationBlock to Program.make.
- Cleaner child OutMessage policies with better inference.
- Better support for optional children with keyed identities.
- Separate guides for migrating from stable and upgrading an earlier preview.
Installation instructions and copyable migration prompts are in the PR description.
github.com/foldkit/foldkit/p…















