惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

B
Blog RSS Feed
量子位
Recent Announcements
Recent Announcements
T
The Blog of Author Tim Ferriss
美团技术团队
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Blog — PlanetScale
Blog — PlanetScale
H
Help Net Security
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
博客园 - Franky
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
宝玉的分享
宝玉的分享
大猫的无限游戏
大猫的无限游戏
V
Visual Studio Blog
博客园 - 聂微东
aimingoo的专栏
aimingoo的专栏
Microsoft Security Blog
Microsoft Security Blog
U
Unit 42
J
Java Code Geeks
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
IT之家
IT之家
Hugging Face - Blog
Hugging Face - Blog
腾讯CDC
L
LangChain Blog

Hacker News: Show HN

PurrrrrFocus: Pomodoro Timer App - App Store Workflow Engine — Multi-Step Orchestration for Bun RapidPhoto: Pro Photo Editor App - App Store GitHub - DheerG/swarms: Achieve extraordinary results with claude code across a variety of tasks SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Show HN: VCoding – A 5 MB native Windows IDE with no dynamic dependencies Show HN: LLMs don't hallucinate because they're bad at math, it's the format GitHub - Agent-FM/agentfm-core: AgentFM is a peer-to-peer network that turns everyday computers into a decentralized AI supercomputer. AgentFM lets you run massive AI workloads directly across a global mesh of idle CPUs and GPUs. Show HN: Tracking Top US Science Olympiad Alumni over Last 25 Years GitHub - Potarix/agent-hub: One place to talk to all your agents Show HN: Runtime security for AI agents(injection,tool abuse, data exfiltration) GitHub - dubeyKartikay/lazyspotify: Terminal Spotify client for macOS and Linux GitHub - the-banana-tool/king-louie: Easy to use GUI Personal AI Assistant. Win/Linux/Mac. Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission GitHub - basteez/jsf-autoreload: maven plugin to enable hot reload on jsf projects uvm32/hosts/host-gdbstub at main · ringtailsoftware/uvm32 GitHub - labsai/EDDI: Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus. GitHub - glitchnsec/fortyone-oss: AI Executive Assistant Platform Quickstart | Alien GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. GitHub - ocrbase-hq/ocrbase: 📄 PDF/IMG ->.MD/JSON Document OCR API for PaddleOCR and GLMOCR. Self-hostable. GitHub - impactjo/home-memory: MCP server that lets your AI assistant remember everything about your home. GitHub - Sets88/dbcls: DbCls is a powerful terminal database client that supports various databases GitHub - neptun2000/heor-agent-mcp GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh RollQuation: Math Puzzles - Apps on Google Play GitHub - dropbox/witchcraft Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis GitHub - opentalon/opentalon: OpenTalon is an open-source platform built from the ground up in Go as a robust alternative to OpenClaw LinkedIn™ 职位抓取工具 - Chrome 应用商店
GitHub - ramayac/GoPOSIX: A Go-native, single-binary POSI...
ramayac · 2026-05-20 · via Hacker News: Show HN

A Go-native, single-binary POSIX with 96.6% BusyBox test compatibility (591/612).

CI go vet coverage Go Report Card Go Version License Docker

Why?

Well, I wanted to do an experiment on Harsness Engineering, and improve my "agentic development" skills, prompts, instructions and all that. I did LFS in my early 20's and I had this weird itch of "do your own thing" but left it alone for my own sanity. Still the POSIX concepts remained in the back of my head. Last year (2025) I started to learn Go-lang, and then LLMs got really good in December 2025. Good enough that I've been using it at work non stop since then.

During that time I got this notion that AI waste time formating output, so I started doing --json output in a lot of my work scripts and tools (to save some time for my robot friends). Eventually all of these random ideas boiled to the conclusion that I should make a complete implementation of POSIX utilities in Go, with a JSON output and a Go SDK, and then benchmark it against BusyBox. It's the "natural conclusion" ... right?

Also deepseek-v4-pro had an very agressive 75% discount until 2026/05/31 15:59 UTC, and I wanted to try pi.dev instead of Antigravity/ClaudeCode.

All things kind of aligned in the last month so here we are now.

I'm not the first to start something like this, there is cugo and go-posix, but sadly they seem to be abandoned, and no wonder! A project like this is a huge undertaking, its probably a year of solid work for 1 human, that being said, took about 3 weeks to do with AI, with the proper "harness" and "agentic development" approach, that's really something.

Anyway the project got into a point that I'm happy with the results, that's enough for me ☑️.

Honest and Obvious Recognitions

I want to be very clear about this:

The only reason this works is that there's a brutally thorough, existing corpus of tests to validate against. The AI iterated until it passed. Without BusyBox's tests, this project is just random hallucinated code. The test suite is the real hero.

  • So check and support BusyBox project and take a look at its amazing test suite, it's a masterpiece of thoroughness and coverage, and it made this project possible.
  • Also Mvdan Shell, it really saved my butt. Absolutely brilliant.
  • And goawk, which I used for the awk implementation, another big save!

Finally: let's not kid ourselves, this project is 90% wiring the AI to do the heavy lifting, 10% is steering it in the right direction, the fact that I was able to "solo dev" this with an LLM, reproducing close to 99% of BusyBox's behavior in a completely different language shows that POSIX utilities are, at their core, text transformers with very well-defined contracts (do one thing and do it well).

Quickstart!

See wiki/sdk.md for the full Go SDK guide and wiki/usage.md for CLI usage and Docker recipes.

CLI (secondary)

# One-shot CLI invocation.
docker pull ghcr.io/ramayac/goposix:cli
docker run --rm ghcr.io/ramayac/goposix:cli ls --json /

Build & Test

make all          # vet + test + build
make test         # unit tests
make testsuite    # BusyBox integration tests (gates every commit)
make ci           # full pipeline (test + testsuite + coverage + docker)

Environment Variables

Variable Default Description
GOPOSIX_SHELL_TIMEOUT 30s Shell execution timeout (Go duration format, e.g. 60s, 5m)

I think there should be more... right?

Performance Highlights

Metric GoPOSIX BusyBox Ratio
Per-call latency (Go SDK, persistent) 60µs 680µs (fork+exec) 11× faster
grep on 100MB file 0.16s 0.86s 5.4× faster (RE2 vs POSIX ERE)
Binary size 10 MB 800 KB 12.5× larger
Cold start 7ms <1ms Not bad, but not great

See Performance Quick Reference and Benchmarking Plan for full details.

Documentation (yes, we have docs and it's decent!)

Quick Project Principles

  • Multicall Binary: Single binary dispatched via symlink or subcommand (goposix ls).
  • Daemon-First: The default image starts the persistent JSON-RPC daemon. Use the Go SDK for programmatic access (60µs/call). CLI is available as a secondary interface (goposix:cli).
  • No CGO: Static compilation for FROM scratch containers (CGO_ENABLED=0).
  • Little Dependencies: Only 3 external Go modules: mvdan.cc/sh/v3 (shell interpreter), golang.org/x/sys (cross-platform syscalls), golang.org/x/term (terminal detection). No external libraries for flag parsing, output, or utility logic.
  • --json Only: Structured output via --json long flag only — no short-form (-j) collision with POSIX flags (ouch!)
  • POSIX Flag Parsing: Custom parser in pkg/common/flags.go with escape hatches for free-form utilities.

Does it work? damn right it does: KoreGoOS.