9885
See what the GitHub community is most excited about today. A bot automatically fetches new repositories from https://github.com/trending and sends them to the channel. Author and maintainer: https://github.com/katursis
#typescript #antigravity_skills #business_knowledge #claude_code #claude_skills #codebase_analysis #codex #codex_skills #developer_tools_ai_agent #gemini_cli_skills #karpathy_llm_wiki #knowledge_base #knowledge_graph #memory #opencode_skills #pi_agent #understandcode #vibe_coding
Understand Anything turns a codebase or docs into an interactive knowledge graph you can search, explore, and ask questions about. It works with tools like Claude Code, Codex, Cursor, Copilot, and Gemini CLI. You can see files, functions, classes, and dependencies in one place, get plain-English explanations, and find what changes affect before you commit. This helps you learn large projects faster, understand how pieces fit together, and save time when onboarding or reviewing code.
https://github.com/Lum1104/Understand-Anything
#csharp #agent_skills
.NET Agent Skills is a set of curated skills and custom agents for coding tools that help with common .NET tasks like coding, data access, debugging, builds, NuGet, upgrades, MAUI, AI, templates, testing, ASP.NET, and new .NET 11 features. You can install them in Copilot CLI, Claude Code, VS Code, Cursor, or Codex CLI. The benefit is faster, more accurate .NET help with less guesswork, so you can build, fix, and improve projects more easily.
https://github.com/dotnet/skills
#jupyter_notebook #gemini #large_language_models #llm #openai #training #transformers
This project shows how to build and train a transformer language model from scratch in PyTorch. It uses the Pile dataset, tokenizes text with tiktoken, and stores tokens in HDF5 files for faster training. The code includes attention, MLP, transformer blocks, training, saving, and text generation. The benefit is that you can learn how LLMs work and train your own small or large model on a single GPU, then use it to generate text for your own tasks.
https://github.com/FareedKhan-dev/train-llm-from-scratch
#javascript #anime #anime_downloader #anime_scraper #downloader #electron #modern_ui #movies #movies_streaming #piracy #series #streaming #streaming_video #tmdb_api #tv
Streambert is a cross-platform desktop app for streaming and downloading movies, TV shows, and anime with no ads or tracking. It also gives subtitles, a library, trending picks, and customization. You can use it to watch and save content faster, keep things organized, and enjoy a more private viewing experience.
https://github.com/truelockmc/streambert
#typescript #ai_agent #ai_coding_agent #anthropic #bun #claude #cli #coding_assistant #llm #mcp #multi_provider #openai #rust #terminal #tui #typescript
omp is a coding agent with the IDE built in. It works on macOS, Linux, and Windows, and gives you many tools for reading, editing, searching, debugging, browser use, and subagents. It can use lots of AI providers and model choices, and it is made to work well right away with real coding tasks. The benefit for you is faster, more accurate coding help in one place, with less setup and fewer extra tools.
https://github.com/can1357/oh-my-pi
#cplusplus
OpenToonz is a free 2D animation program for Windows, macOS, Linux, and BSD. It comes from Studio Ghibli and Toonz Studio, and you can download it, build it from source, or help test and improve it. The main benefit is that it gives you a powerful tool to make animations, with open licensing that allows personal or business use.
https://github.com/opentoonz/opentoonz
#python #agentic_aigc #video_generation
ViMax is an AI tool that can turn an idea, story, novel, or script into a video by itself. It helps with writing the script, planning scenes, choosing reference images, keeping characters consistent, and making the final video. This saves time and makes video creation easier for you, especially if you want to make videos without doing all the technical work.
https://github.com/HKUDS/ViMax
#jupyter_notebook #estimation #perception #robotics #sensorfusion
GTSAM is an open-source C++ library for robot and vision work that helps with mapping and tracking by using factor graphs and Bayes networks. It also has Python and MATLAB support, so you can use it in more than one language. This helps you build more accurate and efficient systems for navigation, SLAM, and IMU data without starting from scratch.
https://github.com/borglab/gtsam
#c_lang #bldc #control #field_oriented_control #foc #hoverboard #motor
This project changes stock hoverboard firmware to use FOC motor control, which can make the motors run more smoothly, more quietly, and with better efficiency, while also allowing higher top speed through field weakening. It supports several control modes and input types, plus safety limits for speed and current. For you, this can mean better ride feel, lower power use, and more flexible ways to build hovercars, boards, robots, or other custom projects.
https://github.com/EFeru/hoverboard-firmware-hack-FOC
#python #ai_agents #amd #comfyui #docker #llama_cpp #llm #local_ai #n8n #nvidia #open_webui #rag #self_hosted #speech_to_text #strix_halo #text_to_speech #workflow_automation
Dream Server lets you run AI on your own machine instead of renting it from a cloud service. It works on Linux, Windows, and macOS, and it can set up chat, voice, agents, search, image tools, and privacy tools with one command. The main benefit is more control: your data stays with you, costs can be lower, and you can keep using AI even without a cloud account.
https://github.com/Light-Heart-Labs/DreamServer
#python #ai #ai_agents #conversational_ai #fastapi #llm #nextjs #open_source #outbound_calls #pipecat #python #self_hosted #speech_to_text #telephony #text_to_speech #voice #voice_agents #voice_ai #voice_assistant #voip #webrtc
Dograh AI is an open-source, self-hostable tool for building voice agents with a drag-and-drop workflow. You can start fast, run it on your own server, use your own LLM, TTS, and STT services, and avoid vendor lock-in. The benefit to you is more control, more privacy, and a working voice bot in minutes without needing API keys.
https://github.com/dograh-hq/dograh
#typescript
CodeGraph helps Claude Code understand a codebase faster by using a local knowledge graph instead of scanning files again and again. It can cut tool calls and time a lot, find symbols, callers, and related code, and keep the graph updated as files change. It works with many languages, stays 100% on your machine, and is easy to start with `npx @colbymchenry/codegraph`. This helps you explore code faster, use fewer tokens, and make changes with more confidence.
https://github.com/colbymchenry/codegraph
#python #automation #claude #mcp #notebooklm #skill
This tool turns many kinds of content, like web pages, PDFs, podcasts, videos, and files, into formats like podcasts, PPTs, mind maps, quizzes, and reports with NotebookLM. It can also handle many sources automatically, including public pages, social posts, and some paywalled articles, and it supports Chinese and English well. The benefit to you is simple: you can save time, understand content faster, and reuse the same material in the format you need for study, work, or sharing.
https://github.com/joeseesun/qiaomu-anything-to-notebooklm
#typescript
# gstack the creator went from 14 lines of code per day in 2013 to 11,417 in 2026—a 810× improvement. It's free, MIT licensed, and works with multiple AI coding agents beyond just Claude.
https://github.com/garrytan/gstack
#typescript #armcord #discord #discord_client #discord_mod #discord_plugin #discord_theme #electron #legcord #nodejs #shelter #vencord
Legcord is a standalone Discord client that blocks all trackers for top privacy, includes built-in mods like Vencord for extras or vanilla use, supports themes, rich presence, and experimental mobile on Linux phones. It's more stable with newer Electron, works cross-platform (Windows, Linux, macOS via easy installs like winget, deb, brew), and avoids the official app. You gain secure, customizable chatting with better performance and no bans reported, despite ToS note—download from legcord.app for instant safe upgrades.
https://github.com/Legcord/Legcord
#python #agentic_ai #agentic_workflow #agents #function_calling #llama_cpp #llamafile #llm #ollama #python #self_hosted #tool_calling
Forge is a Python tool that makes self-hosted LLM tool-calling more reliable. It helps local models handle multi-step tasks with guardrails, better context control, and support for Ollama, llama-server, Llamafile, and Anthropic. You can use it as a workflow runner, middleware, or proxy server with OpenAI-style clients. The benefit is fewer broken tool calls, better results on small models, and easier setup for agent apps, chat tools, and long-running sessions.
https://github.com/antoinezambelli/forge
#go #hysteria #hysteria2 #naive_proxy #shadowsocks #shadowtls #sing_box #trojan #tuic #vless #vmess
S-UI is a web panel for managing SagerNet/Sing-Box. It supports many protocols, several languages, traffic routing, client and system status, and subscription links. It runs on Linux, Windows, and partly on macOS, and can be installed by script, manually, or with Docker. It also supports HTTPS and dark/light mode. This helps you set up and control your proxy service more easily from one place.
https://github.com/alireza0/s-ui
#python #agents #ai #ai_agents #ai_engineering #computer_vision #course #deep_learning #from_scratch #generative_ai #llm #machine_learning #mcp #nlp #python #reinforcement_learning #rust #swarm_intelligence #transformers #tutorial #typescript
This is a free MIT learning guide for AI engineering with 428 lessons in 20 phases. It teaches you AI from the math up, then moves into machine learning, deep learning, LLMs, agents, tools, safety, and production. Each lesson helps you build useful code or AI tools, not just read theory. You can start at the right level, follow a clear path, and keep reusable artifacts for real work. The benefit is simple: you learn how AI actually works and gain practical skills you can use to build and ship better AI systems.
https://github.com/rohitg00/ai-engineering-from-scratch
#typescript #api #gateway #whatsapp
OpenWA is a free, open-source WhatsApp API gateway for developers. It gives you full control, no vendor lock-in, and supports sessions, webhooks, messages, groups, and a web dashboard. It works with SQLite or PostgreSQL, Redis, local or S3 storage, and Docker. The benefit is faster setup, flexible scaling, and secure WhatsApp automation with clear docs and simple API access.
https://github.com/rmyndharis/OpenWA
#go
Files.md is a simple, private app for plain `.md` files. It lets you store notes, documents, journals, tasks, and checklists on your device, with no data sent to a server. You can use it offline, sync later if you want, and even connect it to a chatbot or your own server. The main benefit is clear: you keep full control of your files and can think, write, and organize ideas in one calm place.
https://github.com/zakirullin/files.md
#rust #agentic_coding #ai_coding #anthropic #claude_code #cli #command_line_tool #cost_reduction #developer_tools #llm #open_source #productivity #rust #token_optimization
RTK is a fast command-line proxy for AI tools that cuts token use by 60–90% by cleaning and shrinking command output before it reaches the model. It works with many tools and commands, including Git, tests, builds, and cloud tools, with very little delay. This helps you save money, keep responses shorter, and use AI coding tools more efficiently.
https://github.com/rtk-ai/rtk
#css
PhotoGIMP is a free patch for GIMP that makes it look and work more like Adobe Photoshop, with Photoshop-style tools, shortcuts, and a bigger canvas area. It works on GIMP 3.0 or newer on Linux, Windows, and macOS, and it changes only settings files, not your brushes or plug-ins. The main benefit is that if you know Photoshop, you can start using GIMP faster and feel more comfortable right away.
https://github.com/Diolinux/PhotoGIMP
#batchfile #flighthub #powershell #scripts #tweaks #windows #windows_10 #windows_11 #windows10 #windows11
Awesome Win11 is a guide and script collection for changing Windows 11 with PowerShell and registry tweaks. It helps you adjust the look and behavior of the system, remove or restore apps, change the taskbar, Explorer, and context menu, and disable features like updates, Defender, Cortana, and Edge settings. The benefit is more control, a cleaner desktop, and easier setup without using many third-party tools. Use it carefully, because some changes can break updates or reduce security.
https://github.com/awesome-windows11/windows11
#python
CLI-Anything turns software into agent-ready command line tools, so AI agents can use real apps through simple text commands instead of fragile GUI steps. It supports one-command setup, a 7-step build process, JSON output, REPL use, and a CLI-Hub for browsing and installing tools. The benefit to you is faster automation, less setup work, and more reliable control of software like Blender, GIMP, LibreOffice, and many others.
https://github.com/HKUDS/CLI-Anything
#typescript #agent #ai #antigravity #claude_code #copilot #cursor #skills
Agent Skills is a safe, tested library of skills that adds new powers to AI coding tools like Cursor, Claude Code, and Copilot. It gives you ready-made workflows for planning, cloud help, browser testing, design-to-code, and security review. You can install or update skills with a simple command, and the tool keeps them cached for offline use. This helps you work faster, reduce mistakes, and use AI agents with more trust and control.
https://github.com/tech-leads-club/agent-skills
#jupyter_notebook #agent #agent_framework #agents #ai_agents #deployment #genai #generative_ai #langgraph #llm #llms #mlops #production #python #tutorials
Agents Towards Production is a free open-source guide for building AI agents that work in real products. It gives runnable tutorials on memory, tools, search, deployment, security, monitoring, testing, and user interfaces. You can use it to learn faster, build with less guesswork, and move from a simple prototype to a more reliable, scalable agent system.
https://github.com/NirDiamant/agents-towards-production
#go
witr is a command-line tool that shows why a process, port, file, or service is running. It explains the start chain, source, and context in one clear view, and it can also open an interactive terminal dashboard. This helps you debug faster, understand incidents quickly, and avoid using many separate tools.
https://github.com/pranshuparmar/witr
#c_lang #android #c #ffmpeg #libav #mirroring #recording #screen #sdl2
Scrcpy (v4.0) mirrors your Android device's screen and audio on Linux, Windows, or macOS via USB or Wi-Fi, letting you control it with your computer's keyboard and mouse—no root, app install, ads, or internet needed. It delivers high performance (30-120fps, low 35-70ms latency), recording, camera mirroring, and gamepad support for Android 5.0+. Download only from the official GitHub repo. This saves time, boosts productivity for demos, gaming, or remote control without lag or bloat.
https://github.com/Genymobile/scrcpy
#python #agents #llm #rag #skills #video_analytics #video_search #vlm
NVIDIA's Video Search and Summarization blueprint lets you build AI agents for fast video analysis, including real-time intelligence, alerts, natural language search, Q&A, and long-video summaries using vision models and NIM microservices. Deploy easily via Docker or cloud notebooks on supported NVIDIA hardware. This saves you time analyzing huge video volumes for smart monitoring, warehouses, or operations, boosting decisions and efficiency with ready workflows and custom options.
https://github.com/NVIDIA-AI-Blueprints/video-search-and-summarization
#swift #cpp #csharp #go #ios #java #lightweight #nodejs #on_device #python #rust #swift #text_to_speech #tts #web
Supertonic is a fast, lightweight text-to-speech system that runs directly on your device without needing the internet or cloud services. It supports 31 languages and works across phones, computers, browsers, and other platforms. The system is small enough to run on devices like Raspberry Pi while staying accurate and quick. You get complete privacy since everything happens locally on your device, and you can use it for free with no network dependency. It handles complex text like phone numbers and currency amounts better than many larger systems.
https://github.com/supertone-inc/supertonic