Show HN: GhostBox – Borrow a disposable little machine from the Global Free Tier https://ift.tt/TlBiGMY

Show HN: GhostBox – Borrow a disposable little machine from the Global Free Tier I built this because I was always creating machines on GH actions to test builds on different OS, and I wanted a tight CLI that could do it. I always saw Actions as this great resources and ephemeral machines you could do dev work in just were a natural way for me to work, so this grew out of that workflow. I didn't expect it to blow up, so it wasn't 100% finished when I posted it. But it should stabilize pretty quickly. Happy to know what you think and talk about it. https://ift.tt/H10uD2z May 1, 2026 at 08:22PM

Show HN: N=1 – iOS app for structured longevity self-protocols https://ift.tt/hm0FRdK

Show HN: N=1 – iOS app for structured longevity self-protocols Hello My name is Henry. I built this app for people who want to know for sure that things that they are trying are actually working. I am looking for enthusiastic people who want to see longevity and bio-hacker community grow. At the moment the app is completely free to use. There is no sign up or anything like that. I need your feedback to build something beautiful. https://ift.tt/OTnGEZN May 1, 2026 at 11:00PM

Show HN: Pu.sh – a full coding-agent harness in 400 lines of shell https://ift.tt/MoVyWpG

Show HN: Pu.sh – a full coding-agent harness in 400 lines of shell I originally was just messing with pi-autoresearch. Gave it a sample task to build the most portable coding agent. First cut was 6 KB of shell. Great for one-shots, unusable interactively. I was shocked it actually worked. Started building up -- adding features — but with a self-imposed rule: no new dependencies, and sub 500 LOC. This thing had to be truly portable. Just sh, curl, awk. System primitives only. Which means I did some genuinely disgusting things in awk, including JSON parsing and the OpenAI Responses tool loop with reasoning items carried across turns. It's now ~400 lines. In the box: Anthropic + OpenAI, 7 tools (bash, read, write, edit, grep, find, ls), REPL, auto-compaction, checkpoint/resume, pipe mode, 90 no-API tests. Not in the box: TUI, streaming, images, OAuth, Windows, dignity. Two honest things: 1. I stole/modified the system prompt and the architecture. Pi/Claude/Codex wrote the awk. I cannot read most of this code. This wasn't possible for me a year ago. 2. Heavily inspired by Pi (pi.dev) — same 7-tool surface, same exact-text edit model. Credit where it's due. Pi is awesome -- you should probably use them. The agent loop itself is tiny. Almost everything else in a "real" agent CLI is DX and hardening. You can probably build your own harness exactly how you like it. Mario Zechner's AI Engineer talk on taking back control of your tools nudged me here. The name is because it's a .sh file. The other thing it sounds like is, regrettably, also accurate. https://pu.dev/ May 1, 2026 at 02:25AM

Show HN: Free no-signup site auditor – secrets, subdomain takeover, CVEs https://ift.tt/3bLrPQ0

Show HN: Free no-signup site auditor – secrets, subdomain takeover, CVEs https://ift.tt/hdOPoWc May 1, 2026 at 01:34AM

Show HN: Exploding Hamsters https://ift.tt/HumwgGp

Show HN: Exploding Hamsters https://ift.tt/6cxqH0P May 1, 2026 at 12:20AM

Show HN: An attempt to grow a mind – building software with an inner life https://ift.tt/usiOHYZ

Show HN: An attempt to grow a mind – building software with an inner life https://ift.tt/5D2Utrp April 30, 2026 at 01:27AM

Show HN: Generative UI Library for React https://ift.tt/rnQ65N9

Show HN: Generative UI Library for React https://ift.tt/N7nbaCV April 30, 2026 at 12:58AM