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

J
Java Code Geeks
G
Google Developers Blog
有赞技术团队
有赞技术团队
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Blog — PlanetScale
Blog — PlanetScale
罗磊的独立博客
博客园 - 聂微东
V
Visual Studio Blog
博客园_首页
D
DataBreaches.Net
腾讯CDC
I
InfoQ
F
Fortinet All Blogs
量子位
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
酷 壳 – CoolShell
酷 壳 – CoolShell
博客园 - 【当耐特】
Google DeepMind News
Google DeepMind News
人人都是产品经理
人人都是产品经理
云风的 BLOG
云风的 BLOG
月光博客
月光博客
Recent Announcements
Recent Announcements
MongoDB | Blog
MongoDB | Blog
C
Check Point 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 - DO-SAY-GO/aromatic: Aromatic durable low-bandwid...
keepamovin · 2026-04-25 · via Hacker News: Show HN

Durable store-and-forward transport for unattended devices over Tor

Aromatic is a small demonstrator for moving authenticated low-bandwidth payloads between remote devices and a central collector without public IPs, exposed DNS, or vendor cloud dependency.

It is designed for situations where:

  • networks are awkward
  • topology is opaque
  • public reachability is painful
  • uptime matters more than speed
  • small payloads matter more than streams

Aromatic is not a dashboard, a fleet platform, or a remote shell.

It is a narrow tool with one job: accept small payloads locally, persist them safely, deliver them over Tor, and only forget them after confirmed receipt. That’s it.

Why this exists

A lot of remote infrastructure still depends on fragile assumptions: exposed services, public IPs, VPN complexity, DNS dependency, vendor cloud control planes, and centralized failure domains.

That model works until it doesn’t.

For unattended systems—field sensors, remote telemetry, difficult network environments, long-lived infrastructure—boring durability matters more than convenience theater. Aromatic explores a simpler model:

local process -> durable queue -> Tor -> collector -> durable inbox

No public surface. No required cloud service. No account-based trust model. Just durable delivery.

Architecture

Aromatic is a single Rust binary with two roles:

aromatic node

Runs on the remote device.

  • accepts payloads from local processes
  • stores them durably before transmission
  • sends them to the configured hub over Tor
  • retries if delivery fails
  • deletes only after valid ack

aromatic hub

Runs on the receiving side.

  • receives messages from nodes
  • verifies integrity and identity
  • stores them durably
  • emits ack
  • exposes minimal inspection/export commands

Delivery Semantics (The Actual Product)

Not speed. Not UI. Not dashboards. Durability.

Every message is:

  1. accepted locally
  2. persisted locally
  3. assigned sequence metadata
  4. hashed and signed
  5. transmitted
  6. acknowledged by the hub
  7. removed only after valid ack

If the network disappears, the message stays queued. If Tor restarts, delivery resumes. If the device reboots, the queue survives.

Quickstart

1. Initialize hub

aromatic hub init
aromatic hub invite

2. Initialize node

aromatic node init
aromatic node enroll --invite <code>

3. Send a payload

echo '{"temp_c": -18.2, "battery": 87}' \
  | aromatic node send --stdin --type application/json

4. Inspect hub inbox

aromatic hub inbox

Status & Evaluation Build

Aromatic is a demonstrator. It exists to support real conversations with operators and researchers before pretending we know the final product shape. The goal is not to build a startup from assumptions; the goal is to build less stupid things.

This evaluation build uses shareware-style limits instead of subscriptions or hard shutdowns. Evaluation limits include:

  • 1 hub
  • limited enrolled nodes
  • payload size cap
  • retained message cap

If limits are reached: no silent failure, no data loss, and existing data remains readable/exportable.