15:[[47957127,{"by":"hexer303","descendants":5,"id":47957127,"kids":[47959679,47975156,47960796,47988617,47967807,47958443,47985634],"score":12,"text":"In light of recent news about an agent deleting a production database, I thought now would be a good time to share this.

As the use of AI tools in production is becoming more common, sadly so will the high profile incidents like the one mentioned.

Fewshell is a terminal agent specifically designed to avoid this.

There is no setting to enable command auto-approval. This is by-design, so that the user never has to second-guess or worry about accidentally having it enabled.

Originally my intention was to build an AI mobile terminal to make typing shell commands easy. But with so many mobile-enabled 'claw' agents being available, I decided to make Fewshell the opposite of an autonomous agent.

Please star if you like, let me know what you think. Happy to answer questions.

About me: I'm an ex Amazon Sr. SDE for Alexa AI, and currently am working in AI safety research for agentic RLVR. I use this tool to run and check on my lab experiments.","time":1777513784,"title":"Show HN: Agent that refuses to run commands without human approval","type":"story","url":"https://github.com/few-sh/fewshell"}],[47932974,{"by":"pizza_man","descendants":18,"id":47932974,"kids":[47937209,47965763,47957973,47944323,47935592,48085945,48084153,47946384,47944996],"score":34,"time":1777375754,"title":"Show HN: I wrote a DOOM clone in my own programming language","type":"story","url":"https://spectrelang.org/log/devlog#cubedoom"}],[47916180,{"by":"rabbithols","descendants":8,"id":47916180,"kids":[47923342,47919438,47945612,47919813],"score":10,"text":"Most PDF editors just overlay text boxes. CrabPDF modifies the actual text inside the file. Nothing leaves your computer... no server, no account, works offline.","time":1777247994,"title":"Show HN: CrabPDF – privacy-first PDF editor that edits real text","type":"story","url":"https://crabpdf.com/"}],[47912662,{"by":"zachdotai","descendants":4,"id":47912662,"kids":[47912663,47912739],"score":3,"time":1777228572,"title":"I built an agent that breaks your AI agents before someone else does","type":"story","url":"https://fabraix.com/"}],[47904167,{"by":"jmt710","descendants":4,"id":47904167,"kids":[47908914,47906044,47904789,47912697,47908398],"score":6,"text":"Describe a task, and AI will give you a breakdown of whether it can do your task well, poorly, or somewhere in between. I built it mostly because I kept getting asked "what is AI even good for" and fumbling the answer.

The most fun use is testing it on things you already know it can't do and seeing how it explains why it can't be done.","time":1777147679,"title":"Show HN: Good AI Task – a tool for asking AI what it can and can't do","type":"story","url":"https://goodaitask.com"}],[47873966,{"by":"damiannn","descendants":70,"id":47873966,"kids":[47875156,47875115,47875961,48037434,47900840,47875271,47918938,47948355,47905323,47874200,47877789,47875549,47885565,47874957,47875150,47875234,47903307,47875820,47875326,47875893,47875852,47875402,47874818,47874989,47875680,47875757,47873967,47876231,47876839,47876001],"score":74,"text":"Hey folks,

I built Hisorty, a small daily game where you sort historical events in chronological order (strongly inspired by Wordle mechanics).

Each day you get 6 events and 3 attempts to place them correctly.

I found it surprisingly tricky to design those puzzles so the difficulty is just about right.

Would love feedback on:\n- difficulty balance\n- UI/UX \n- whether the concept feels engaging long-term

Happy to answer any questions about how I built it.","time":1776939286,"title":"Show HN: Built a daily game where you sort historical events chronologically","type":"story","url":"https://hisorty.app/"}],[47859099,{"by":"morawr","descendants":7,"id":47859099,"kids":[47880464,47861106,47869744,47868119,47860337],"score":11,"text":"I built a plugin that lets Claude Code delegate work to Gemini CLI.

I started this after finding myself reaching for Gemini more often on long context repo work. I have been especially liking Gemini’s codebaseinvestigator for long context.

This is inspired by openai/codex-plugin-cc.

Code Review, adversarial review. Under the hood its Gemini CLI over ACP

Would love feedback from people using Claude Code, Gemini CLI, or ACP. I am especially curious whether this feels useful outside my own workflow.

Its a great combo with Opus 4.7 + Gemini 3.1 workflows","time":1776832869,"title":"Show HN: Gemini Plugin for Claude Code","type":"story","url":"https://github.com/m-ghalib/gemini-plugin-cc"}],[47865594,{"by":"npilk","descendants":9,"id":47865594,"kids":[47879219,47866629,47866972,47867988,47872609,47869647,47869242],"score":5,"text":"$16","time":1776873966,"title":"Show HN: A free tool for non-technical folks to easily publish a website","type":"story","url":"https://weejur.com"}],[47828968,{"by":"juanpabloaj","descendants":12,"id":47828968,"kids":[47830002,47829726,47833272,47832410,47831985,47834214,47829627,47831006,47829697,47865115,47831489,47830292,47850385,47850543,47843899,47831323,47829845,47836735],"score":58,"time":1776644484,"title":"Show HN: A lightweight way to make agents talk without paying for API usage","type":"story","url":"https://juanpabloaj.com/2026/04/16/a-lightweight-way-to-make-agents-talk-without-paying-for-api-usage/"}],[47853551,{"by":"informal007","descendants":2,"id":47853551,"kids":[47863097,47864081],"score":5,"time":1776800560,"title":"Show HN: Humanoid.js – One HTML file that scores how human your clicks look","type":"story","url":"https://humanoid-js.bensontech.dev/"}],[47834659,{"by":"g_33_k","descendants":9,"id":47834659,"kids":[47837313,47834660,47854373,47835469,47835796],"score":7,"time":1776694385,"title":"Show HN: Self-hosted Raspberry Pi wall display (no cloud, no subscription)","type":"story","url":"https://github.com/silentg33k/chalkboard-installer"}]]

user:	natloz
created:	April 21, 2026
karma:	2