GitHub Trending Today #55: free-my-tv, pele, SPACE, ascii-sharks, harp, DeepInfer, ElyOffice, fx0

Author: Github Awesome · Watch on YouTube

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: GitHub Trending Today #55: free-my-tv, pele, SPACE, ascii-sharks, harp, DeepInfer, ElyOffice, fx0 · Channel: Github Awesome · 15:00 · Watch on YouTube

Topics: AI agents, Software engineering with AI, Local and open-source models, Privacy and data protection

Summary

This episode of GitHub Trending Today presents 35 open-source projects, spanning TV customization, developer tools, AI workflows, databases, and graphics. It highlights practical features such as local processing, compatibility with existing tools, and privacy-conscious design, while also covering agent-based coding and research applications. Several projects focus on performance, including Rust-based systems, speculative decoding, and GPU acceleration. The video concludes with a computer vision library designed as an alternative to OpenCV.

Key points

Free my TV offers an ad-free Google TV and Android TV home screen, while TV detox and Dumb TV provide tools to simplify or customize television setups.
Pelle and Andean render Mermaid diagrams as SVG, with different approaches to rendering and build integration.
Space provides isolated agent-code sandboxes using Rust and Firecracker microVMs; Harp and Pandacode organize coding-agent workflows.
Several tools apply AI to specific tasks, including local video editing, handwritten tutoring, spoken screen recordings, and paper-based coding exercises.
Deep Infer and Capybara focus on model inference, while Eagle 4 and Mem T 1.0 explore speculative decoding and persistent model memory.
Rubio and KiviDB are Rust-based Redis-compatible data stores, with KiviDB also offering search indexes and clustering features.
Developer utilities include hot-reloading Rust code, multi-computer desk sharing, and reviewing code changes through quizzes.
The episode also features interface frameworks and utilities, including Nanoui, NSH, and Drag Shift.
Solid CV is a Rust computer vision library aiming to replace OpenCV without its C++ build chain.

Chapters

Start	Chapter	Summary
00:00	TV interfaces, diagrams, and agent sandboxes	The episode introduces Free my TV, Pelle, Space, ASCII Sharks, and Harp, then covers Eagle 4 and animated terminal art.
03:06	Office apps, security, and AI-assisted work	The projects include Euly Office, Deep Infer, a spyware scanner, local video editing, an AI tutor, TV cleanup, and the FX0 design agent.
05:42	Coding workflows and developer tools	The video covers Pandacode, Scroll, Andean, market-data skills, Rust hot reloading, multi-computer sharing, and Superlight.
08:49	Writing, databases, memory, and local models	This section introduces GrammAE, Redis-compatible servers, Mem T 1.0, TV customization, Capybara, and paper-based coding exercises.
11:24	Voice assistants, code review, and app frameworks	The projects include an ESP32 voice assistant, KiviDB, Sphinx, Omacast, and the Nanoui and NSH interface frameworks.
13:59	File tools and computer vision	The closing projects are Drag Shift, which edits files during drag-and-drop, and Solid CV, a Rust computer vision library.

Quotes

Free my TV is an ad-free home screen for Google TV and Android TV. — 00:00

each one gets its own VM and file system. — 01:01

visitors receive finished diagrams without downloading the parser and layout engine. — 06:45

English and German checks run inside your browser with bundled dictionaries — 09:19

Solid CV is a computer vision library written entirely in Rust aiming to replace OpenCV without the C++ build chain. — 14:29

Concepts and entities

Free My TV, Pele, SPACE, ASCII Sharks, Harp, Eagle 4, DeepInfer, ElyOffice, Zero Breach, Editany, FX0, PandaCode, Scroll, Trader skill, Superlight, Rubio, MemT-1.0, Capybara, KiviDB, Sphinx, Solid CV

Full transcript (original language)

00:00 Welcome back to GitHub awesome. This is GitHub trending today number 55. 35 trending open source projects on GitHub right now. Let’s get into it. Free my TV is an ad-free home screen for Google TV and Android TV. Built as a plain Android app of about 2 megabytes. Your streaming apps sit on a glass style shelf. The home screen shows three trending picks each hour, taken only from apps you already have, along with an hourly forecast and a big clock

00:31 screensaver. The theme switches between day and night at your city’s sunset. Pelle is a small library from the Obsidian team that turns mermaid diagram text into SVG with no runtime dependencies. It covers every diagram type in mermaid 12.1, but it uses its own layout engine and visual style. Call render and you get the SVG back right away with no waiting. Colors and fonts come from CSS variables, so a theme change updates without redrawing anything.

01:01 Space is Perplexity’s open source platform for running agent code safely, written in Rust on Firecracker microVMs. Your app makes an HTTP call to create a sandbox from a template, and each one gets its own VM and file system. From there, you run commands, stream the output back, and pause, suspend, or wake the sandbox whenever you need it. It works with the E2B Python SDK, so existing code can move over. ASCII Sharks puts swimming sharks and fish on a browser canvas drawn entirely

01:32 with text characters. You choose the shark species and sizes, then let its TypeScript engine handle animation and resizing. Clicking creates a splash. There are no runtime dependencies, and it pauses when the canvas leaves the screen. It also respects reduced motion settings by showing a still frame. Harp is an unofficial fork of the open source Warp terminal, made for running a lot of Claude code and Codex agents at once. It swaps the tab bar for a sidebar where each row shows whether that agent

02:04 is working, waiting for you, or done. You can sort tabs into named groups, and with tmux turned on, they survive a restart. A footer shows your 5-hour and weekly usage limits as rings. Eagle 4 is the newest version of the Eagle speculative decoding method for speeding up large language models. A small draft model guesses the next several tokens, and the big model checks them all at once. So, the output stays the same as before. It plugs into SG Lang and vLLM through

02:36 engine hooks, and sizes its guest tree from measured costs on your hardware. ASCII magic art lets you browse animated text art right in your terminal, then copy a prompt that helps your coding agent add it to a website. You can switch the artwork into dots, binary, or your own characters, and embed it with a web component. Each piece is also an editable JavaScript module. Web animations pause offscreen and show a still frame when visitors prefer reduced motion.

03:06 Euly Office wraps established document engines in desktop apps built with Rust and GPU. Word processing, spreadsheets, and presentations use LibreOffice Kit, while its PDF app adds editing and OCR through MuPDF and Tesseract. Your files stay in ordinary folders with no account or cloud service required. There are Markdown and Git apps, too. Deep Infer is DeepSeek’s open-source inference framework, the same one that runs its production services built for DeepSeek v4.1 flash. It started out

03:37 built on vLLM and runs on DeepSeek’s own kernels like Flash MLA, Deep Gem, and DPP. It supports expert parallelism for mixture of experts models, speculative decoding, and structured JSON output on both Nvidia and Huawei Ascend chips. Zero Breach is a read-only scanner that checks your computer for traces of the Ad fection spyware campaign, which researchers have tied to Candiru. It’s plain Python with no dependencies, and it blocks its own network access while

04:08 it runs. It compares your browser history, caches, and certain files against 338 published indicators. Every finding gets graded by how strong the evidence is. Editany is a skill for Claude code and Codex that edits video locally when you describe what you want. Open your agent in a folder of footage and explain the edit. It transcribes the speech on your own machine, picks complete takes, and checks each cut against the waveform. It also removes dead air, breaths, and

04:39 repeated lines. You review a short sample before it renders the final MP4 with HyperFrames. Are pairs an AI tutor with a pen and paper you’re already using. Describe what you want to learn, work through its exercises, then submit your handwritten work for feedback through a camera or photo upload. It saves progress and reflections locally to help adapt later lessons. The app runs on your computer, but submitted work and relevant learning context go to OpenAI

05:09 for tutoring. TV detox speeds up an old, slow Android TV by having Claude code or Codex clean it up over Wi-Fi. Paste its prompt into your agent, and it lists the TV’s apps and asks before switching anything off. Nothing gets deleted, and it keeps an undo list as it goes. Core parts like Google Play services and settings are on a protected list it never touches. Then it installs a tiny home screen that uses about 30 megabytes of RAM. FX0 is Vercel Labs open-source design

05:42 agent built on its lib effects agent library. Describe a page in chat, and the agent writes it as one self-contained HTML file. Each design gets version history, so you can copy or download any earlier version. The agent can search the web for references and work in one of 10 bundled styles or one you upload. What stands out is that every step is saved as it runs. Pandacode puts the coding agents you already use into one desktop workspace alongside Git diffs, a terminal, and a

06:12 browser. Choose a different agent and model per task, track work on a board, and review each turn’s changes before committing. Separate Git work trees let tasks use their own checkouts. Conversations stay local while prompts go to your chosen provider. You still need its CLI and account. Scroll lets you show an AI agent a problem instead of typing it out. Built in Rust with Swift helpers for macOS. Press a shortcut, then talk and draw red marks on your screen while it records.

06:45 Press it again, and Scroll transcribes your voice on the device and picks key frames from moments like here or a click. It packs a contact sheet, full frames, and a time-aligned transcript into one bundle for your agent. Andean compiles Mermaid diagrams into SVG before your page reaches the browser. Its Rust compiler plugs into Vite or Markdown builds, so visitors receive finished diagrams without downloading the parser and layout engine. You can also render files from the command line

07:15 or generate SVG on a server. Colors follow CSS variables, so the same compiled diagram can switch between light and dark themes without another build. The Trader skill is a set of 11 Claude skills for working with public market data, built as plain Python with no dependencies. One reads an investor’s quarterly SEC filings and shows what changed in their portfolio. Others track news, congressional trade disclosures, and earnings, size positions to your budget, and check each

07:46 order against your risk limits. Rajit is an experiment that hot reloads Rust code in milliseconds using Rust analyzer as the front end and Cranelift for the machine code with no LLVM and no linker. When you save and edit, it recompiles only the functions that actually changed and swaps them in live, and your program state stays where it was. The demo is Conway’s Game of Life where changing the rules shows up on the very next frame. Twin desk links two computers on the same network so they work like one desk

08:18 across Mac, Windows, and Linux. It’s built on Electron with a small native helper for each system and WebRTC for the screen video. One mouse and keyboard move across both machines. You can extend or mirror a screen, share the clipboard, drag files from one to the other, and open the other computer’s apps in their own windows. Superlight is a workflow plugin for Claude code that takes a spec and comes back with finished, tested work built on the ideas of the Superpowers plugin. It

08:49 plans the build and splits it into tasks that run in parallel waves with up to 18 agents at once. Simple work goes to Haiku, implementation to Sonnet, and the final review to Opus. A fresh reviewer then reports problems without editing anything. GrammAE is a free writing assistant for Chrome that checks spelling and grammar as you type. English and German checks run inside your browser with bundled dictionaries so you don’t need an account or an API

09:19 key. It underlines problems where you’re writing and you click to accept or dismiss each fix. It also catches mixed-up words that are spelled correctly but used wrong. Rubio is a key-value server written in Rust that speaks the Redis protocol so redis-cli and your existing Redis clients connect without changes. Each key belongs to one shard, which handles its requests without locks or thread switches. Every write goes to a log before it’s applied. Then come the built-in commands like

09:51 leases with fencing tokens, semaphores, capped counters, versioned rights, and delayed queue entries. Mem T 1.0 is a memory model from Mem tensor that builds lasting memory right into the model. Built on Qwen 3.6 and Deep Seek V4 flash. A memory block sits next to the model’s attention and keeps a memory state that updates after every turn. On the next turn, the model reads from that state instead of re-reading the whole conversation. The memory stays the same

10:22 size no matter how long the history gets. Dumb TV strips a supported Fire TV or Android TV back to your apps and HDMI inputs using ADB without root. Its own launcher has no internet permission, while cleanup scripts disable tracking packages and unwanted services. It snapshots your settings first, so a restore script can put them back. Support varies by model, and streaming apps you choose to uninstall need reinstalling from the store. Capybara is a fast engine for running

10:53 Qwen and Neumotron models locally on Apple Silicon Macs. It reads your prompt on both the GPU and the neural engine. Then it writes the answer with speculative decoding, where a small draft guesses ahead and the main model checks the guesses. One command starts a chat page plus OpenAI and Anthropic compatible APIs, and it can launch Claude code or Codex against your local model. Paper lead turns AI research papers into coding exercises you can run on your own

11:24 CPU. Read the method, fill in its core Python or PyTorch functions, then check your implementation against mathematical properties and gradient tests. Interactive labs show how your changes affect the results with deliberately buggy examples to explain common mistakes. SP32 thinks turns a small ESP32S3 gadget, the Sensecap Watcher with a round screen and a knob, into a voice assistant built on Meta’s open-source Muse Gadget SDK. Say “Hi ESP” or press

11:54 the knob and ask a question. The device trims your recording, sends it off for transcription, and streams the reply back as speech one sentence at a time. Speech-to-text, chat, and text-to-speech can each come from a different provider. KiviDB is a Redis-compatible in-memory database written in Rust, so your existing Redis clients connect without changes. It splits data across shards, saves snapshots and an optional append-only log, and uses io_uring on Linux when the kernel allows it. It also

12:25 includes Redis stack-style search with full-text, tag, numeric, and vector indexes. Clustering, replication, and access control come built in. Sphinx pauses a Git push to quiz you on what your code changes actually do. A separate reviewer model reads the committed diff without seeing your coding conversation, then asks about behavior and failure cases. Miss something and it explains the issue before giving you another scenario. You can answer with the code open using your own model provider.

12:57 Omacast is a Raycast-style app launcher for Omar key, the Linux setup built on Hyperland. It runs as a plugin inside Omar key’s own shell with no electron and no extra background services. Press Alt and space, and one list searches your apps, about 400 Omar key menu actions, and the web. Results get ranked by how often and how recently you use things, and it learns which words you type for each one. Nanoui is a framework for building desktop apps in Zig. Each app either

13:27 draws its own native UI on the GPU or shows a webpage in the system web view. It ports the core of Egoist’s MyGo and borrows ideas from Zed’s GPUI like chain styling and a virtualized list. The native views rebuild from your app state on each event. On macOS, they’re drawn with Metal and an idle app uses no CPU at all. NSH lets you write interfaces in Rust, TSX, or both and renders them through each platform’s own UI system. TSX screens can reuse your Rust components

13:59 with hot updates that preserve application state. You can embed it inside an existing native app or package an app on its own, including a static website. Hosts can also load signed TSX packages at runtime with explicitly configured sandbox capabilities. Drag Shift is a Windows utility for converting and editing files in the middle of a drag. Start dragging a file, press Shift, and a radial menu pops up. Drop the file on a slice to convert it, like an image to PDF or a PDF to Word.

14:29 Press Shift again for tools such as compress, resize, crop, redact, or strip GPS data. The result is saved next to the original, which stays untouched. Solid CV is a computer vision library written entirely in Rust aiming to replace OpenCV without the C++ build chain. It covers the familiar pieces like filters, edge detection, ORB features, stereo matching, and optical flow with kernels tuned for Apple silicon and Nvidia GPUs. It also adds tools for vision language models like

15:00 skipping image patches that haven’t changed between frames.