German iGerman00

Installing Tailscale on GL.iNet SiFlower (SiChang) routers

I have no experience creating opkg packages for OpenWRT, so manual installation it is!

Automatic installation

Lanrat's Tailscale Repository

It's not compressed as of writing this, but it's much easier to install, and now supports our architecture

Manual compilation and installation

Prerequisites

A GL.iNet SiFlower router, like the Opal (GL-SFT1200)

Floatplane YouTube Hotkeys

This is a relatively simple, decent script to introduce some extra hotkeys to Floatplane. By default, only some hotkeys are active.
I've only tested this script on my Chromium-based browser with Tampermonkey, however I have no reason to believe it won't work on other browsers.

Hotkeys that I have added:

k, MediaPlayPause - play/pause
StopMedia - pause
j - seek 10 seconds back
l - seek 10 seconds forward

DIY Private Filtered DNS

Create your own secure DNS server with filtering capabilities
NextDNS, eat your heart out

This tutorial will guide you through setting up a private DNS server using Caddy and AdGuard Home. You'll create a secure, encrypted personal DNS endpoint with content filtering and authorization that you can use from anywhere in the world.

What you'll get

A personal DNS server that blocks ads and unwanted content
Encrypted DNS connections for privacy

Riffusion system prompts

2025-07-21

To the best of my abilities, I have extracted the system prompts used by Riffusion. LLMs can be seen used in the two main modes: "Prompt mode" and "Compose mode". The LLM they use is unknown. This kind of functionality has existed on Riffusion since their launch (2025-01-30). Their prompts mention reasoning, so it might be a chain of thought model, given the timeline -- something like o1-mini, but I have not been able to confirm this. For all I know, Riffusion being hosted on Google, it might be one of the Gemini models.

Failure on title generation (occurs normally, sometimes) might help realize what model or framework is being used:

}}]}]}]}}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]}]} to=functions.final_result payload={title:

Definitely seems to be using function calling-capable models. Maybe PydanticAI on the backend?

	U_XqO[287568] = (function () {
	var p$m = 2;
	for (; p$m !== 9; ) {
	switch (p$m) {
	case 2:
	p$m = typeof globalThis === "\x6f\u0062\u006a\u0065\u0063\x74" ? 1 : 5;
	break;
	case 1:
	return globalThis;
	break;

	import { escapeHtml } from './escape.js';
	import template from './template.html';

	async function listAllKeys(namespace, options = {}) {
	let allKeys = [];
	let cursor = undefined;
	let listComplete = false;

	while (!listComplete) {
	const currentOptions = { ...options, cursor: cursor, limit: 1000 };