Burrow Tunnel vs DClipVideo

Burrow is a frp client built for developers and homelab tinkerers. It lets you securely connect from your phone to your own Mac, Linux, NAS, or GPU machine at home, and keep doing the things you would normally do in front of a computer. [Why Burrow] The local AI agent era is here. You might be running Claude Code, Cursor agents, local LLMs, Stable Diffusion, or other self-hosted services on a workstation, Mac mini, or NAS at home. When you are out, you want to use your phone to check progress, tweak configs, transfer files, run maintenance scripts. Exposing ports directly to the public internet is dangerous. Commercial remote-desktop tools route your traffic through their servers. Burrow uses frp's stcp visitor mode: - End-to-end encrypted; the key stays only between you and your devices - No public ports exposed on your home machines - All traffic flows only through your own frps server, with no third-party relay - We cannot, and never will, see any of your data [Three access modes] 1. Shell (PTY terminal) A full SSH terminal with support for vim, htop, tmux, color output, and ANSI escapes. Password and public-key authentication. Built-in favorites and runbooks: pin commands like "restart service", "tail logs", or "kick off training run", and execute them in one tap. 2. SFTP (file browser) Browse the remote filesystem by directory. Upload, download, rename, delete. Useful for pulling trained model weights, generated images, or logs back to your phone. 3. HTTP (in-app browser) A built-in WebView for accessing intranet web services: Jupyter, Open WebUI, Home Assistant, Grafana, the frps dashboard, your router admin page, self-hosted photo libraries, and so on. Self-signed certificate trust supported. [Who Burrow is for] - People running AI agents, local LLMs, or self-hosted services at home who need safe remote access - Developers who want to do server ops, check logs, or run scripts from their phone - Homelab enthusiasts and self-hosters who already run their own frps and at least one Linux or macOS machine [Who it is not for] - Users who want zero-config, install-and-go (you need to understand frp configuration) - Anyone needing graphical remote desktop (Burrow only does command line, files, and web) - Anyone expecting the app to run continuously in the background (limited by iOS foreground model) [What you need before starting] 1. A frps server you have deployed (a VPS, or another machine on your network) 2. A frpc daemon already running on the home machine you want to reach, with an stcp-type proxy configured 3. The same sk (shared key) on both sides 4. SSH enabled on the remote if you want shell, or a web service running if you want HTTP The "User Guide" section inside the app has a full getting-started walkthrough with example configs. [Privacy] Burrow does not collect any user data. All connection configurations (addresses, passwords, keys) are stored only on your device, in the iOS Keychain. No accounts. No login. No analytics. No third-party SDKs. [Feedback] This is the work of a solo developer. There is a feedback channel at the bottom of the in-app User Guide. If something does not work for you, please tell me.

### English ClipVideo - Scientific Intensive Listening System ClipVideo implements a proven five-step training methodology: Import → Sentence Loop → Speed Drill → AB Reinforce → Stats Feedback. Unlike conventional media players, ClipVideo deconstructs video content into its smallest learnable unit — the sentence. By controlling playback speed and setting precise repeat intervals, you can systematically conquer listening challenges. Visual learning analytics quantify your progress, empowering you to optimize your training strategy. Key Features: • Content Management: Import and organize local videos as learning materials • Sentence-Level Training: Intelligent segmentation with single-sentence loop • Speed Drills: Continuous 0.5x-2.0x adjustment for progressive difficulty • AB Intervals: Precise start/end markers for targeted weak-point reinforcement • Learning Dashboard: Daily/weekly/monthly charts to quantify training output • Offline Architecture: Local SQLite storage with zero network dependency ClipVideo — not a media player, but your listening training laboratory.