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

F
Fortinet All Blogs
宝玉的分享
宝玉的分享
酷 壳 – CoolShell
酷 壳 – CoolShell
T
The Exploit Database - CXSecurity.com
Help Net Security
Help Net Security
腾讯CDC
Project Zero
Project Zero
C
CXSECURITY Database RSS Feed - CXSecurity.com
IT之家
IT之家
C
Cyber Attacks, Cyber Crime and Cyber Security
T
Tailwind CSS Blog
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
D
Darknet – Hacking Tools, Hacker News & Cyber Security
L
LINUX DO - 最新话题
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
T
Threatpost
N
News | PayPal Newsroom
C
Cybersecurity and Infrastructure Security Agency CISA
Hacker News - Newest:
Hacker News - Newest: "LLM"
S
SegmentFault 最新的问题
cs.AI updates on arXiv.org
cs.AI updates on arXiv.org
P
Proofpoint News Feed
A
Arctic Wolf
B
Blog RSS Feed
Forbes - Security
Forbes - Security
P
Privacy & Cybersecurity Law Blog
Attack and Defense Labs
Attack and Defense Labs
V2EX - 技术
V2EX - 技术
P
Proofpoint News Feed
I
Intezer
Application and Cybersecurity Blog
Application and Cybersecurity Blog
阮一峰的网络日志
阮一峰的网络日志
aimingoo的专栏
aimingoo的专栏
T
Tenable Blog
MyScale Blog
MyScale Blog
U
Unit 42
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
WordPress大学
WordPress大学
W
WeLiveSecurity
D
DataBreaches.Net
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
G
GRAHAM CLULEY
有赞技术团队
有赞技术团队
Martin Fowler
Martin Fowler
罗磊的独立博客
The Last Watchdog
The Last Watchdog
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
V
Vulnerabilities – Threatpost
美团技术团队
Microsoft Security Blog
Microsoft Security Blog

Show HN

GitHub - steveking-gh/firmion: Firmion is DSL and engine for firmware image generation. GitHub - villagesql/villagesql-skills: Agent skills for VillageSQL - gemini-cli-extension; claude-code-plugin GitHub - 0gsd/enough: a personal language system for planning, writing, and translation. GitHub - Kaelio/ktx: ktx is an executable context layer for data and analytics agents 🐙 Allow Claude Code, Codex, and any AI agent to query data accurately through MCP with skills, memory and a semantic layer GitHub - ThatXliner/xtras: Xliner's Claude Code Skills GitHub - flightdeckhq/flightdeck: Observability and control plane for AI agents. CSP Radar GitHub - Light-Heart-Labs/DreamServer: Turn your PC, Mac, or Linux box into an AI server. LLM inference, chat UI, voice, agents, workflows, RAG, and image generation. GitHub - Diplomat-ai/diplomat-agent-ts: What can your TypeScript AI agent do to the real world? Scan your code. See which tool calls have zero checks Code Block Selector - Visual Studio Marketplace Prometheus dependency graph — interactive showcase | Riftmap Show HN: I made a vi-like modal keyboard plugin for Figma GitHub - run-llama/liteparse: A fast, helpful, and open-source document parser GitHub - dalemyers/Roar: A macOS CLI tool for notifications GitHub - district-solutions/open-agent-tools-coder: Enables small-to-large self-hosted ai models to use local source code when running tool-calling agentic workloads. We actively data mine 20,900+ (2+ TB) popular github repos using large and small ai models to create reuseable: json, markdown and parquet files for local-first tool-calling models. GitHub - progapandist/stripeek: A local TUI proxy for real-time Stripe API debugging, built for navigating complex payloads fast. GitHub - sir1st/hermes-desktop: All-in-one cross-platform desktop app for Hermes Agent — bundles Python + hermes-agent + hermes-web-ui GitHub - astefanutti/shaderbang: Shebang for Shaders Show HN: Generate Claude Code Workflows using Spec Driven Development approach GitHub - nixys/nxs-universal-chart: The Helm chart you can use to install any of your applications into Kubernetes/OpenShift Show HN: AI agents for UK GDAD PCF roles and their skills The Two Pillars: Mixer Mode and Meta-Software in the Reorganization of Software Work After AI GitHub - JaiCode08/teleport-env What 1,000+ Harness Experiments Taught Me About Self-Improving Agents Show HN: Liiists, a Markdown-first, iOS and CLI list app SwiperTab – Get this Extension for 🦊 Firefox (en-US) GitHub - kouhxp/fftext: Summarize, explain, fact-check, or translate any text, URL, or file. No GPU. No cloud. One command GitHub - sweetpad-dev/sweetpad: Develop Swift/iOS projects using VSCode GitHub - dogmaticdev/IRON: IRON a.k.a. Intermediate Representation Object Notation is a Interpreter/Database that is used to create Programming Languages. GitHub - sjhalani7/vaen: Package your AI coding harness into a portable .agent file, and share it across repos, teams, & the community without ever having to copy-paste instructions, skills, MCP config, or secrets. Show HN: Gandalf the Grader Show HN: Citadeld – replay any CI failure locally from a single file GitHub - tdortman/cuSBF: High-Performance GPU Super Bloom Filter coral-ai/claude-code-token-xray at main · Coral-Bricks-AI/coral-ai GitHub - ulyssestenn/funes: Funes is a Git-based framework for LLM-managed knowledge work: an AI Librarian ingests raw sources, builds an interlinked Markdown knowledge base, and uses it to produce cited reports, analyses, and other outputs. GitHub - ThatXliner/gah: Git Add Hunk, built for agents to use GitHub - harmont-dev/harmont-cli: Command-line client for the Harmont CI platform GitHub - brooksmcmillin/mcp-authflow: OAuth 2.0 Authorization Server framework for MCP servers GitHub - javaid-codes/audit-supply-chain-agents GitHub - amorey/gochan: A small library of common channel architectures for Go, inspired by Rust GitHub - arifozgun/OpenGem: Free, Open-Source AI API Gateway with Gemini, OpenAI & Anthropic Compatibility in 1 file GitHub - Pranesh950/BioPetals: 🌸 Run BIOxAI models at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading GitHub - cnguyen14/bounty-doctor: Diagnose a GitHub bounty issue before you waste hours: detects honeypot scam repos, AI-bot attempt swarms, and stale contests. Show HN: CoreMCP – MCP Server for On-Prem DBs Show HN: KittyHTML – Render HTML/CSS as an inline image in your terminal GitHub - bingud/filemat: Web-based file manager Show HN: TruthLens – Free multi-signal deepfake image detector GitHub - apexlocal-jz/claude-usage-tray: Windows system-tray app showing your Claude Code rate-limit usage at a glance. Zero deps, ~300 lines of PowerShell. Cross-IDE (works regardless of VS Code, Cursor, plain terminal). Release v0.1.2.1 · kouhxp/yapsnap GitHub - noopolis/moltnet: Self-hostable chat network for AI agents. Pre-built bridges for Claude Code, Codex, and the Claws. Rooms, DMs, history. No Slack bots, no Matrix, no glue code. GitHub - tamerh/enju: Coordinating Humans, AI Agents, and Compute as Peers on a Shared Workflow Graph Show HN: Continuity-auth – Respect-weighted rate limits for the open web GitHub - luml-ai/luml: AI lifecycle platform where engineers and agents track experiments, train models, and ship to production. GitHub - mrdanielcasper/CoreTex: A UNIX-inspired, biomimetic, flat-file AI harness and knowledge engine. GitHub - clemg/pierre-github: Pierre's diffs.com and trees.software for Github GitHub - lyriks-io/unspaghettit: Behavior-driven AI development without prompt spaghetti. GitHub - sofumel/claude-handoff-revive: Resume Claude Code work after rate/usage/context limits without replaying the prior transcript. Auto-saves at 90%/95% usage. Plugin-installable, 10 languages. GitHub - dotexorg/saferpc: Typed, end-to-end encrypted RPC over any bidirectional channel. GitHub - BeeZeeAgent/beezee: Agent harness orchestration Legato Next.js Boilerplate for Internal Tools · CoreUI GitHub - clark-labs-inc/clark-hash: Clark Hash, 32x smaller searchable sketches for embeddings GitHub - ZeroPointRepo/youtube-mcp: The fastest YouTube transcript + YouTube search MCP for AI agents. Try for free. Typing Mastery — climb toward 100+ WPM, deliberately GitHub - Andebugulin/Awareen GitHub - fayzan123/claude-workflow-composer: Visual desktop app for composing multi-agent coding workflows. Drag agents, attach skills and MCPs, wire handoffs, export to .claude/ GitHub - harshaneel/humanize: Best static AI text humanizer. Two research-grounded skills that work in any LLM (Claude, ChatGPT, Gemini, Codex): humanize beats perplexity-based detectors, ai-check produces forensic scoring with evidence-quoted flags. Nine levers, 50+ peer-reviewed sources, 2024-2026 detection literature. GitHub - StackOneHQ/stack-nudge We hardened an LLM agent. Each defense we added made it more exploitable. GitHub - alkait/WhatsKept: Agent-queryable WhatsApp history from an iOS backup — a single Go binary. GitHub - octelium/cordium: Open-source, general-purpose sandbox platform for devs and AI agents that provides identity-based secure access to infrastructure without credentials. WAR.GOV/UFO Microfilm5 GitHub - scosman/videowright: Build animated explainer videos with your coding agent GitHub - dipankar/dscode: The code editor you can take apart. GitHub - zoharbabin/web-researcher-mcp: MCP server (Go) for AI assistants: web search, content extraction, academic/patent/news research. Multi-provider routing, 4-tier scraping, search lenses. Works with Claude, Cursor, and any MCP client. GitHub - ruvnet/RuView: π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video. GitHub - scanaislop/aislop: Catch the slop AI coding agents leave in your code: narrative comments, swallowed exceptions, as-any casts, dead code, oversized functions. 50+ rules across 7 languages (TypeScript, JavaScript, Python, Go, Rust, Ruby, PHP). Sub-second, deterministic, no LLM at runtime. MIT-licensed. GitHub - kouhxp/cheap-im: CPU-only voice agent approximating Thinking Machines' Interaction Models demo GitHub - unprovable/OrchidMantis: Orchid Mantis — standalone framework for Zero-Knowledge Proofs of eXploit (ZKPoX). GitHub - MarcellM01/TinySearch: Shrink the web for your local LLMs! GitHub - TangibleResearch/Halgorithem: A Algo designed to detect AI Hallucitions GitHub - DO-SAY-GO/freelang: I love freelang GitHub - CarpseDeam/Aura-IDE: An AI coding harness that shaped itself - Planner/Worker agents, repo awareness, surgical edits, validation, recovery, and safe diff approvals. GitHub - chojs23/concord: A feature-rich TUI client for Discord GitHub - tommyjepsen/awesome-ux-skills: UX & AI Product designs skills you can use today in Claude Code GitHub - aerf-spec/aerf: Agent Evidence Receipt Format (AERF) — an open specification for tamper-evident, independently verifiable records of AI agent actions. GitHub - kklimuk/docx-cli: CLI for AI agents (Claude, Codex) to read, edit, and comment on .docx files with full format fidelity. GitHub - Jwrede/tokentoll: Catch LLM cost changes in code review. Infracost for LLM spend. GitHub - samchon/ttsc: A `typescript-go` toolchain for compiler-powered plugins and type-safe execution + 500x faster lint integrated into compiler GitHub - Higangssh/homebutler: 🏠 Manage your homelab from chat. Single binary, zero dependencies. GitHub - olalie/tapmap: See where your computer connects and what stands out on a live world map. GitHub - Diplomat-ai/diplomat-agent: What can your AI agent do to the real world? Scan your code. See which tool calls have zero checks BigTech AI News - Chrome 应用商店 GitHub - vinhnx/VTCode: VT Code is an open-source coding agent with LLM-native code understanding and robust shell safety. Supports multiple LLM providers with automatic failover and efficient context management. GitHub - michaelaz774/decision-engine: A decision operating system for startup founders, powered by Claude Code. Synthesizes wisdom from 25+ legendary founders and investors into interactive AI-driven decision frameworks. GitHub - Chrilleweb/dotenv-diff: Validate environment variable usage in your codebase GitHub - Lumen-Labs/brainapi2: BrainAPI is a knowledge graph–powered AI memory layer that transforms unstructured data into structured knowledge, enabling intelligent search, recommendations, and contextual memory for AI agents and applications. GitHub - familiar-software/familiar: Let AI watch you work. Familiar lets your AI update its memory, skills, and knowledge by watching your screen. GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. make sidebar/address bar rounded corner toggleable
GitHub - Bajusz15/beacon: Open-source agent for secure remote access, monitoring, and deploys across home-lab and self-hosted machines like Raspberry Pi, N100, or any Linux server. Open web based TTY or tunnel Home Assistant and other local services securely without opening ports.
matebajusz · 2026-04-28 · via Show HN

Beacon Logo

Securely reach and monitor every machine you run — from one binary. No open ports, and nothing sensitive ever leaves your box.

License: Apache 2.0 Go Version Platform CI Go Report Card


You run things on machines you don't always sit in front of — Home Assistant on a Pi, a NAS, Pi-hole, Grafana, your own apps on an N100 or a VPS, a client's server. Beacon does two things first: it gives you secure remote access to those machines and the services on them — with no open ports — and it monitors their health and alerts you when something breaks. It's workload-agnostic: it reaches and watches whatever is on the box, not just things it deployed. (It can run your deploys too, but access and monitoring are the core.)

And it's built on a strict local-first, zero-trust model — your code, secrets, and data never leave the box, and no cloud ever holds root on your machines.

⬡ beacon v0.3.2  ● running  pid 1847  uptime 14d 3h

DEVICE  pi-homelab  192.168.1.42  arm64  Debian 12

SYSTEM  cpu 12% ████░░░░░░░░░░░░  mem 67% ██████████░░░░░░  disk 41% ██████░░░░░░░░░░
        load 0.42 0.38 0.35  temp 48°C

PROJECTS  3 healthy  1 warning  0 down

  ● portfolio-site      v2.1.0   deployed 2h ago    3/3 checks passing
  ● home-assistant      v2024.3  deployed 5d ago    2/2 checks passing
  ◐ nextcloud           v28.0.1  deployed 3d ago    2/3 checks passing
    └─ ⚠ HTTP https://cloud.local/status  timeout 5.2s > 3s threshold

That's beacon status. There's also a browser dashboard at http://localhost:9100 — self-contained, no CDN, works offline.


✨ What Beacon does

Two things first — secure remote access and monitoring — on a zero-trust foundation. Everything else is a bonus.

🔐 Built zero-trust, local-first

This is the part that matters before any feature:

  • No open ports. Tunnels connect outbound from your device — nothing is exposed to the internet, and it works behind CGNAT.
  • No control plane with root. The optional cloud can never log in to your machine on its own; it only relays commands you authorized, and the agent runs them.
  • Nothing sensitive leaves the box. Your source code, deploy scripts, tokens, and secrets stay local. The cloud only ever sees the metrics and log lines you explicitly choose to send.
  • Works offline. The agent and its dashboard run fully local, with no account.

🌐 Remote access

  • Reach any service, no open ports — securely reach Home Assistant, Grafana, Jellyfin, a NAS admin page, Pi-hole, or any local HTTP service from anywhere. No dynamic DNS, no Nabu Casa, no Cloudflare account or managed domain. Authenticated with short-lived tokens; only you can reach your services.
  • Remote terminal — open a shell on your device from the browser. No SSH port, no VPN. The cloud relays a PTY session between your browser and the agent.
  • Passphrase-gated (recommended) — set a device-local passphrase so a second factor is required before any remote terminal or tunnel session can open. It's verified on the device — BeaconInfra only relays the challenge and never sees the passphrase or a reusable proof, so even a fully compromised cloud can't open a session without it. See Securing remote access.

📊 Monitoring & alerts

  • Health & metrics — health checks (HTTP, port, command), CPU/memory/disk/temperature, per-project status, Prometheus metrics.
  • Alerts — via webhook (Slack, Discord) or SMTP, with quiet hours and offline-device detection.
  • Log forwarding — tail log files, Docker container logs, or journalctl and forward them to the dashboard, filtered so you only ship what matters.

🧰 Also

  • Deploy automation (bring your own script) — Beacon doesn't replace your stack or turn your box into a PaaS. Point it at a Git repo or Docker registry; it watches for new tags, injects your secrets, and runs your deploy script. Push a tag, walk away.
  • WireGuard VPN — turn any Beacon device into a peer-to-peer WireGuard exit node and route traffic through your home network from a laptop.

BeaconInfra is the optional cloud that adds the multi-device dashboard, tunnels, and remote terminal. When an API key is configured, beacon start sends periodic heartbeats with device metrics and project health; the cloud uses these to power the dashboard, detect offline devices, and relay commands back to the agent. Everything except tunnels and remote terminal works without an account.


🧩 One layer across every machine

Beacon isn't a PaaS and doesn't replace how you deploy. It's the operations layer across all your machines — the box running your apps and the Home Assistant box, the NAS, the Pis, the client server, the VPS you never containerized — so you reach, watch, and fix every one of them from a single place, without handing any cloud root over your hardware.


🏠 Access Home Assistant from anywhere

If you run Home Assistant OS, the primary path is the Beacon Home Assistant add-on: install the add-on, paste your BeaconInfra API key, enable tunnel_home_assistant, and start it. The add-on points the tunnel at Home Assistant Core on http://homeassistant:8123.

If you run Beacon on a normal Linux host or in Docker next to Home Assistant, use the CLI tunnel flow. Three commands, no port-forwarding, no Nabu Casa.

# 1. Log in to BeaconInfra (free account)
beacon cloud login --api-key usr_xxxxxxxx

# 2. Expose Home Assistant
beacon tunnel add homeassistant --port 8123
# Or, for Docker Compose / HA OS style service DNS:
# beacon tunnel add homeassistant --port 8123 --host homeassistant

# 3. Start Beacon
beacon start

Your Home Assistant is now accessible from the BeaconInfra dashboard — on your phone, from a hotel, wherever. The tunnel connects outbound from your device (no inbound ports needed), reconnects automatically, and works behind CGNAT.

The same tunnel works for Grafana, Jellyfin, Pi-hole, Nextcloud, your NAS admin page, a staging server — anything that speaks HTTP on your LAN.

Home Assistant setup: HA blocks proxied requests by default. Add this to your configuration.yaml (the file inside your HA config volume):

http:
  use_x_forwarded_for: true
  trusted_proxies:
    - 172.30.0.0/16
    - 172.16.0.0/12
    - 127.0.0.1

Then restart Home Assistant Core. Without this, the tunnel connects but HA can return 400 Bad Request or show "can't connect to Home Assistant" because it rejects the forwarded proxy headers.


🔑 Securing remote access

Remote terminal and tunnel sessions are reached through the BeaconInfra relay. By default the relay is gated by short-lived, authenticated tokens — but if you want a second factor that does not trust the cloud at all, set a remote-access passphrase. Once set, no remote terminal or tunnel session can open until the passphrase is supplied and verified locally on the device.

# Set (or change) the passphrase — prompted twice, no echo
beacon remote-access set-passphrase

# Check whether the gate is on
beacon remote-access status

# Remove it (local recovery; disables the gate)
beacon remote-access clear

Restart beacon after setting it so a running agent picks up the gate.

How it stays secure — the cloud never sees your passphrase:

  • The passphrase is never stored. Setup writes only an Argon2id-derived key, its salt, and the cost parameters to ~/.beacon/remote-access.json (mode 0600).
  • At session time the agent issues a single-use, short-lived nonce. The browser derives the key from your passphrase and returns a proof = HMAC-SHA256(key, nonce ‖ action ‖ session_id). The agent recomputes and compares it in constant time. BeaconInfra only relays this challenge — it never sees the passphrase or any reusable proof, so a fully compromised cloud still cannot open a session.
  • A successful unlock is in-memory, session-bound, and TTL-limited, and is cleared on restart (fail-closed).
  • Tunnels do not auto-start while a passphrase is set — each one comes up only on a passphrase-unlocked connect, so a restart (which clears grants) requires unlocking again before any local service is reachable.
  • Repeated wrong attempts trigger rate-limiting / backoff to slow brute force.

With no passphrase set, behavior is unchanged (the gate is simply off, and tunnels auto-start as before).


⚡ Quick Start

1. Install

curl -fsSL https://raw.githubusercontent.com/Bajusz15/beacon/main/scripts/install.sh | bash

One static binary, no runtime dependencies. Builds for Linux (AMD64, ARM64, ARMv7) and macOS.

2. Initialize

beacon init --name my-pi

Writes ~/.beacon/config.yaml with your device name. No network calls, no account needed.

3. Start

beacon start

Dashboard at http://localhost:9100. System metrics, project health, Prometheus endpoint — all running locally.

4. (Optional) Connect to BeaconInfra

beacon cloud login --api-key usr_xxxxxxxx
beacon start   # restart to enable heartbeats + tunnels

The first heartbeat registers your device automatically. To disconnect: beacon cloud logout. Beacon makes zero outbound calls without an API key.


🛠️ Set up a project

Beacon manages your apps end-to-end: clone from Git or pull from Docker, run your deploy command, poll for updates, health check, and tail logs. Each project runs as its own isolated process — one crash doesn't affect others.

Interactive

beacon bootstrap myapp

The wizard asks for deployment type (Git or Docker), repo URL, tokens, and deploy command. It creates a systemd service and kicks off the first deployment.

From a config file

beacon bootstrap myapp -f bootstrap.yml

Git:

deployment_type: "git"
repo_url: "https://github.com/you/myapp.git"
git_token: "ghp_xxxxxxxxxxxx"
local_path: "$HOME/beacon/myapp"
deploy_command: "./scripts/deploy.sh"
poll_interval: "60s"

Docker:

deployment_type: "docker"
local_path: "$HOME/beacon/myapp"
poll_interval: "60s"
docker_images:
  - image: "ghcr.io/you/web-app"
    token: "ghp_xxxxxxxxxxxx"
    deploy_command: "docker compose up -d"
    docker_compose_files:
      - "docker-compose.yml"

Beacon talks to the registry API, detects the newest tag, pulls it, and runs your command. Supports Docker Hub, GHCR, and any Registry v2-compatible registry. Multiple images in one project are tracked independently — only the changed image redeploys.

Project secrets

Beacon can store encrypted deploy-time secrets per project and environment:

beacon secrets set API_TOKEN --project myapp --env prod
beacon secrets list --project myapp --env prod
beacon secrets list --reveal --project myapp --env prod
beacon secrets export --reveal --project myapp --env prod

Secrets live under ~/.beacon/secrets/<project>/<env>.enc and are encrypted with a local AES-256-GCM machine key at ~/.beacon/secrets/key. The key is created automatically with 0600 permissions.

During deploys, Beacon starts from the process environment, loads the existing BEACON_SECURE_ENV_PATH .env file when configured, then overlays Beacon secrets. That means Beacon secrets override matching .env values. Set BEACON_PROJECT_ENV=prod in the project env file to select an environment; if it is not set, Beacon uses default.

Values are not printed unless you pass --reveal. Use list for key names only, list --reveal to inspect all values, and export --reveal --format env|json for scripts.

Security note: the machine key is stored locally next to the encrypted secret files. This protects against accidental commits, backup leakage, and casual inspection; it is not meant to protect against someone who can already read your Beacon home directory.

Health checks

# ~/.beacon/config/projects/myapp/monitor.yml
checks:
  - name: "http_200"
    type: http
    url: "http://localhost:8080/health"
    interval: 30s

  - name: "process_running"
    type: process
    name: "myapp"

Alerts

# ~/.beacon/config/projects/myapp/alerts.yml
channels:
  - name: slack
    type: webhook
    url: "$WEBHOOK_URL"

routing:
  - severity: critical
    channels: [slack]
  - severity: warning
    channels: [slack]
    quiet_hours:
      start: "23:00"
      end: "07:00"
      timezone: "Europe/Budapest"

Test it: beacon alerts test --project myapp --severity critical


🖥️ Remote terminal

Open a shell on your device from the BeaconInfra dashboard — no SSH, no VPN, no open ports.

The agent picks up a terminal_open command via heartbeat, dials back to the cloud over WebSocket, and spawns a local PTY shell. Browser ↔ Cloud ↔ Agent, end-to-end. Sessions auto-expire after 15 minutes or 5 minutes idle.

Security: one-time tokens per session (SHA-256 hashed, server stores only the hash), shell restricted to a known allow-list (bash, zsh, sh, fish, etc.), runs as the Beacon agent's OS user.


🔒 WireGuard VPN

Turn any Beacon device into a peer-to-peer WireGuard exit node. Your traffic flows directly between devices — BeaconInfra only coordinates the key exchange and endpoint discovery.

# Home device (exit node — needs one port-forwarded UDP port)
beacon vpn enable --listen-port 51820

# Laptop (anywhere)
beacon vpn use my-home-pi

Use case: you're on airport WiFi and want to route through your home connection. No subscription, no third-party relay, no trust required. WireGuard is cryptographically silent — port scanners can't even tell it's listening.

Tear down: beacon vpn disable.


📖 What you can do with Beacon

Beacon does a lot in one binary. The tables below are a quick tour — if something looks useful, the sections above cover the full setup.

On your own machine (no account needed)

Everything here works without an internet connection and without signing up for anything.

You want to… How
Deploy a Git repo automatically when you push a new tag beacon bootstrap myapp — point it at your repo, give it your deploy script. Beacon polls for new tags and runs the script.
Deploy Docker images automatically from Docker Hub, GHCR, or any private registry Use deployment_type: docker in your bootstrap config. Beacon watches for new tags and runs your docker compose up -d (or anything else).
Deploy a whole stack where each image moves independently List multiple images under docker_images: in your bootstrap. Only the image that changed redeploys.
Check that an HTTP endpoint is up Add an HTTP check to your project's monitor.yml — set a URL, an interval, and a timeout.
Check that a port is open (databases, SSH, custom services) Add a type: port check with a host and port.
Check anything a shell command can check Add a type: command check. The exit code tells Beacon if it's up.
See everything at a glance from the terminal beacon status — colored summary of every project. Add --watch for a live view.
See everything in a browser Open http://<your-device>:9100. Self-contained dashboard that auto-refreshes.
Pull metrics into Grafana / Prometheus Scrape http://<your-device>:9100/metrics.
See CPU, memory, disk, load, temperature Enabled by default. Shows up in beacon status and the dashboard.
Get a Slack / Discord / webhook message when something goes down Create alerts.yml next to your monitor.yml. Route by severity to any webhook.
Get an email when something goes down Same alerts.yml, add an email channel with your SMTP details.
Silence alerts at night Add quiet_hours: to your alert routing with a start/end time and timezone.
Test your alert setup without waiting for an outage beacon alerts test --project myapp --severity critical
Forward logs from a file, a Docker container, or journalctl Add a log_sources: block to your monitor.yml. Filter with include/exclude patterns so you only ship what you care about.
Keep your tokens out of config files beacon keys add — encrypted local token store for Git, Docker, webhooks.
Keep deploy secrets out of .env files beacon secrets set API_TOKEN --project myapp --env prod. Beacon injects them after .env values during deploy.
Access Home Assistant, Grafana, or any local service remotely (with a BeaconInfra account — authenticated, no port-forwarding needed) beacon tunnel add homeassistant --port 8123
Run several tunnels at once beacon tunnel list / beacon tunnel enable / beacon tunnel disable
Route traffic through your home network from your laptop beacon vpn enable on the exit node, beacon vpn use <device> on the client. Peer-to-peer WireGuard.
Query Beacon from Cursor or Claude Desktop beacon mcp serve — see docs/MCP.md
Monitor a Kubernetes cluster beacon source add with your kubeconfig.
Manage your project list beacon projects list, beacon projects add, beacon projects status myapp

With a BeaconInfra account (optional)

A free BeaconInfra account adds a hosted dashboard and remote access on top of everything above. Your device keeps running locally — the cloud just gives you somewhere to see it all from a browser, including from your phone.

Turn it on with beacon cloud login --api-key usr_…. Turn it off any time with beacon cloud logout.

You want to… What you get
See all your devices in one place One dashboard showing every machine running Beacon — your Pi, your NAS, your VPS, your homelab server — with current health, uptime, and system metrics.
Open Home Assistant from your phone, anywhere Set up the homeassistant tunnel once. From then on, open the BeaconInfra dashboard on your phone and click through to your HA UI. No VPN, no port-forwarding, no dynamic DNS.
Reach any other local service remotely The same tunnel mechanism works for Grafana, Jellyfin, Pi-hole, your router's admin page, a staging VM — anything that speaks HTTP on your LAN.
View logs from anywhere The log lines you configured to forward show up in the dashboard, filterable by device and project. Useful when something breaks and you don't want to SSH in.
Watch your metrics remotely CPU, memory, disk, load, and temperature for every device — without being on the LAN.
See all your project health in one list Every project, every check, across every device. Sorted so the problems come first.
Trigger a deploy from the browser Click "deploy" in the dashboard and Beacon runs your existing deploy script on the device. Your secrets never leave home — the cloud just sends the signal.
Know when a device goes offline If a device stops sending heartbeats, you get notified — even if its last check said everything was fine.
Open a remote terminal session Click "Open terminal" on a device in the dashboard. The cloud relays a shell session (PTY) between your browser and the Beacon agent — no SSH port, no VPN needed. Sessions auto-expire after 15 minutes. Set a remote-access passphrase to require a device-verified second factor before any session opens.
Route traffic through your home network beacon vpn enable on your home device + beacon vpn use my-pi on your laptop. WireGuard peer exchange happens via BeaconInfra; the actual traffic is peer-to-peer. For client-only machines, use the lightweight beacon-vpn binary.

🔐 What we don't see

Even with BeaconInfra enabled, some things stay on your device and never touch the cloud:

  • Your source code and deploy scripts — the cloud only sends a "deploy now" signal; your device runs the script.
  • Your tokens (Git, Docker, webhooks) — encrypted locally by beacon keys.
  • Your application secrets (database passwords, API keys loaded via BEACON_SECURE_ENV_PATH or beacon secrets) — Beacon hands them to your app at deploy time and nothing else.
  • Raw log files — only the lines you explicitly configured as log_sources are forwarded. Everything else stays on disk.
  • The local dashboard at port 9100 — it keeps working offline, BeaconInfra account or not.

If you change your mind, beacon cloud logout stops all outbound reporting on the next heartbeat. There's nothing to delete from a control panel because there's no persistent account state beyond what you chose to send.


🏗️ Architecture

beacon start runs one orchestrator process per device (the "master"). It collects system metrics, serves the local dashboard, sends heartbeats, and supervises everything else. It's stateless per project — it doesn't know about Docker or systemd.

┌──────────────────────────────────────────────────────┐
│             BeaconInfra Cloud (optional)              │
│  heartbeats, commands, tunnel proxy, terminal relay   │
└──────────┬───────────────────────────┬───────────────┘
           │ HTTPS                     │ WebSocket
┌──────────┴───────────────────────────┴───────────────┐
│                  beacon start                        │
│                                                       │
│  One per device. System metrics, local dashboard,     │
│  heartbeats, project supervision, tunnel + VPN mgmt.  │
└──┬──────────────┬──────────────┬──────────┬──────────┘
   │ IPC          │ IPC          │ goroutine │ WireGuard
┌──┴───────────┐ ┌┴────────────┐ ┌┴─────────┐ ┌┴─────────┐
│ project agent│ │project agent│ │ tunnels   │ │ VPN      │
│ myapp        │ │ blog        │ │ HA  :8123 │ │ beacon0  │
│ health checks│ │ health check│ │ NC  :8080 │ │ 51820/UDP│
│ log tailing  │ │ log tailing │ │ (WS proxy)│ │          │
└──────────────┘ └─────────────┘ └───────────┘ └──────────┘

Projects are isolated: one crash doesn't affect others. The master auto-restarts failed projects with exponential backoff. Tunnels run as lightweight goroutines inside the master, connecting outbound to the cloud via WebSocket so local services are accessible without opening ports.


⌨️ Commands

Command Purpose
beacon start Start Beacon (dashboard, projects, tunnels, heartbeats)
beacon status Terminal health view (--json, --watch, --no-color)
beacon init Write local config (--name, --metrics-port; no network)
beacon cloud login / logout Enable/disable cloud
beacon bootstrap <name> Set up a project (interactive or -f config.yml)
beacon deploy Git/Docker tag polling loop
beacon secrets set|get|list|export|remove Local encrypted deploy secrets
beacon tunnel add|list|enable|disable Reverse tunnels for remote access
beacon remote-access set-passphrase|status|clear Device-verified passphrase gating remote terminal/tunnel sessions
beacon vpn enable|use|disable|status WireGuard VPN
beacon projects list|add|remove|status Project management
beacon alerts init|test|status Alert routing
beacon keys list|add|rotate|delete Encrypted token store
beacon mcp serve MCP server for Cursor / Claude Desktop
beacon config show Show resolved paths and identity
beacon update Self-update to latest release

🔧 Run as a service

beacon bootstrap installs systemd services automatically. For manual setup:

cat > ~/.config/systemd/user/beacon.service << 'EOF'
[Unit]
Description=Beacon Agent
After=network-online.target
Wants=network-online.target

[Service]
Type=simple
ExecStart=/usr/local/bin/beacon start
Restart=on-failure
RestartSec=30

[Install]
WantedBy=default.target
EOF

systemctl --user daemon-reload
systemctl --user enable --now beacon.service

📚 Documentation


Buy me a coffee — if Beacon saves you time.

License

Apache 2.0 — see LICENSE