Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
EmptyDrum
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
1.
▲
by
EmptyDrum
7mo ago
That's not the point of the skill.
2.
▲
by
EmptyDrum
7mo ago
I agree - in the end we have some fuzzy turing machine when writing in pseudo code. I think it is inherently hard to get this right. I also think that these decisions about control flow etc should stay with the developer given I find it ver
3.
▲
by
EmptyDrum
7mo ago
I feel like that this is fundamentally impossible to solve for. Approximately the effort = planning + checking correctness seems to be constant.
4.
▲
by
EmptyDrum
7mo ago
Spam
5.
▲
by
EmptyDrum
7mo ago
https://github.com/HalfEmptyDrum/Pseudo-Code-Flow the link again
6.
▲
by
EmptyDrum
7mo ago
Btw, I got inspired by this: https://www.williamjbowman.com/blog/2026/03/05/against-vibes... Really useful read!
7.
▲
by
EmptyDrum
7mo ago
True (will fix) - but then still, Claude will still not change my pseudo code, but directly translate. I liked that part particularly.
8.
▲
Maybe we can keep on coding? pseudo code project
9 points
by
EmptyDrum
7mo ago
|
12 comments
9.
▲
Press-One: Auto-accept every Claude Code prompt
(github.com)
7 points
by
EmptyDrum
7mo ago
|
2 comments
10.
▲
by
EmptyDrum
7mo ago
I kept mashing 1 to accept changes in Claude Code, so I automated it. It spawns your command in a PTY and pipes "1" into stdin on a loop. npm install -g press-one press-one --delay 3000 claude "fix all the bugs"
11.
▲
by
EmptyDrum
7mo ago
Yes, I partially used their approach which I found very smart. Added a few improvements which you can find on the GitHub.
12.
▲
Show HN: Context-compact – Summarize agent context instead of truncating it
(github.com)
6 points
by
EmptyDrum
7mo ago
|
2 comments
13.
▲
by
EmptyDrum
7mo ago
fyi - this was built inspired by how OpenClaw manages llm-keys. I liked their idea and improved on it. Feedback is welcome.
14.
▲
Show HN: key-carousel - Key rotation for LLM agents
(github.com)
5 points
by
EmptyDrum
7mo ago
|
2 comments