explaingit

Popular Python repositories on GitHub, each explained for non-coders — what it does, its tech stack, and how to run it.

• sky3100/frciblyk12A Chinese language web panel for creating, tracking, and configuring free ChatGPT accounts at scale.
• ingar30/reviewerA multi-agent Python tool that automates peer review of academic economics papers, producing one consolidated
• romarayt/raytsystem-public-osA local-first, self-hosted workspace giving one person or a small team a command center for documents, tasks,
• sgloria/research-paper-pipelineAn AI pipeline that automates writing an academic research paper end to end, from idea generation through a su
• baoxin1100/ok-kesA Windows automation tool for the game Chaos Zero Nightmare that uses image recognition to auto-battle, manage
• happy-chen-ch/text_classificationA Chinese news headline classifier comparing traditional ML, FastText, BERT, and a distilled small model acros
• liuluhaixiu/davinci-autoedit-agentA Codex skill that automates editing raw video footage into a finished cut inside DaVinci Resolve.
• roboparty/ufoA research framework that trains humanoid robots, mainly the Unitree G1, to move using unsupervised reinforcem
• toolsai/grok-build-connectorAn Agent Skill that lets an AI coding assistant hold a live, visible conversation with xAI's Grok model for se
• vita-epfl/rdmResearch code for training AI image generators that produce a finished image in a single step, matching a publ
• xinhuangcs/agentmakerA Python framework for building AI agents with memory, tool use, and document search built in, working with mo
• a2328275243/mempalace-evolveA Python library that gives AI agents persistent, self-cleaning memory across sessions using vector search and
• anthonykhayesaudsrx50512/flash-usdt-senderA Python tool that generates crypto transactions showing a temporary, non-final balance on wallets and block e
• confluentinc/quickstart-streaming-agentsA Confluent Cloud quickstart with four hands-on labs showing how to build real-time AI agents on Apache Kafka
• django-haystack/saved_searchesA Django add-on that records what each user searches for on your site, letting you build features like recent
• donalleniii/lingbot-desktop-macA Mac app that turns a dragged-in video into an explorable 3D point cloud, running the Lingbot-Map reconstruct
• onlyterp/prompt-cache-skillsA collection of drop-in fixes that turn on and correct prompt caching in popular AI coding agents like Cline,
• stainlu/stainfulOpen-source Python SDK generator that reads an OpenAPI 3 spec plus a stainless.yml config and emits a hand-wri
• albertcheng19/medskillosAn open-source framework that adds structured medical knowledge, safety boundaries, and evidence tracking to A
• amap-ml/roleagentA research framework adding two reward-shaping techniques to train AI agents on multi-step text tasks.
• gongnyang/gn-voiceA Claude Code skill that rewrites AI drafted Korean text to match one author's real writing style, verified wi
• krishnaik06/multiple-linear-regressionA beginner-friendly learning resource that explains multiple linear regression — predicting one outcome from s
• nalltama/raivA Windows image viewer that automatically AI-upscales and denoises images as you browse, using Real-CUGAN or R
• noz2/roothoundRoothound reads LinPEAS scan output and builds a visual graph showing every possible path from your current us
• role1776/netmonA self-hosted bot that monitors your internet speed and counts devices on your home network, then sends you Te
• tencent-hunyuan/hy-mt2Tencent's family of fast-thinking translation models (1.8B, 7B, 30B-A3B MoE) covering 33 languages, with on-de
• ymsniper/bbfBBF is a command line tool that brute-forces an unknown part of a Bluetooth device's hidden address, for autho
• google-cloud-ai/alphaevolve-on-googlecloudA Python library and examples for running AlphaEvolve, Google's Gemini powered agent that evolves and scores c
• humanmllm/swimOfficial CVPR 2026 code for SWIM, a video MLLM fine-tuned from Qwen2.5-VL-7B that aligns attention maps with r
• igorbarshteyn/jlens-ggufAn interactive tool to visualize and steer what's happening inside a GGUF language model as it generates text,
• jingyi-wu-richael/rachel-digital-human-productionA Codex AI skill that automates producing an authorized digital-human talking-head video using MiniMax voice c
• juew/skill-routing-kitA local-first Codex plugin that improves skill and plugin routing accuracy so the assistant picks the right ca
• justxor/pythonroamap2026A free, Russian-language roadmap and companion course that walks a learner from Python basics to senior-level
• noietch/eva-clientA browser based framework that deploys, evaluates, and collects data from real robots, working across multiple
• xmu-xiaoma666/x-dreamerAn AI tool that generates detailed, photorealistic 3D models from text descriptions by borrowing knowledge fro
• ltczding-gif/ref-downloaderref-downloader takes a paper's DOI or PDF and automatically downloads every reference it cites as a PDF, using
• simonw/micropython-wasmA Python package that runs small MicroPython code snippets inside a sandboxed WebAssembly container, with no f
• somnusochi/vlm-autoyoloA web platform that auto-labels images and video with AI based on a text description, then trains a YOLO objec
• wasasquatch/comfyui_viewer_openreel_extensionA ComfyUI extension that embeds the OpenReel video editor so you can trim, add effects, and add text overlays
• zengweishuai/scalebfmScaleBFM is the in-progress code release for a research paper on a foundation model for humanoid robot behavio
• dcdmllm/instructsamA research model that segments individual objects in an image from a plain-language description, category name
• gmboquet/mixleA Python library that automatically figures out and trains statistical or neural models from messy, mixed data
• palaiologos1453/openinterviewA locally-run AI mock interview tool for Java backend and AI development jobs, with a 700+ question bank, scor
• slavasexton/comfyui-ocioAdds professional Nuke-style color management to ComfyUI, letting users read, grade, and write color-accurate
• thiagofinch/mega-brainA local-first knowledge system for Claude Code that turns your videos, PDFs, and notes into a searchable AI br
• yuecheng919/gemdepthA research model that estimates how far away objects are in video footage while keeping depth stable and consi
• auucoder/gptgrok2apiA self hosted gateway that turns GPT and Grok subscription accounts into an OpenAI-compatible API, bundled wit
• derekh-233/amdlA Windows desktop app with a dark web-style interface that wraps the open-source gamdl engine to download song
• juneyaooo/nihaixiaAn AI assistant skill that searches a large archive of Ni Haixia's Traditional Chinese Medicine course materia
• kanna12580/kk-knowledge-agentA lightweight knowledge base tool that stores and searches your documents by meaning, exposed as an MCP tool A
• legenddriftragemystic6648212/ph-content-download-2026A Windows desktop app that downloads videos from adult content sites like PornHub by pasting in a URL and pick
• mattj85/spookiuiA terminal app that lets you browse and edit every Ghostty terminal setting with live preview, without hand ed
• wan-video/wan-skillsPlug-in skills that let AI assistants generate images and create PowerPoint slide decks on their own, powered
• gao-pooh/paper-scraperA Python tool that searches and batch-downloads academic PDFs from ScienceDirect and INFORMS using your instit
• lingxiaoyiyu-hub/grok-register-winA Windows and Linux tool with a web panel that automates signing up for Grok (xAI) accounts using disposable e
• modelscope/flowraFlowra lets developers wrap machine learning models and data-processing code into reusable visual blocks that
• primeintellect-ai/experiments-autonomous-speedrunningA raw research archive showing two AI agents, Claude Code and OpenAI Codex, independently competing to optimiz
• wanshuiyin/aris-in-ai-offerBilingual ML interview cheat-sheet pack on LLMs, diffusion, RLHF, agents, and MoE, with foundations, 25 tiered
• weicj/vllm-2080ti-definitiveA fork of vLLM tuned to run large AI language models fast on a pair of older, affordable RTX 2080 Ti GPUs link
• zju-real/sdarA research codebase that trains AI agents to complete multi-step tasks faster and more stably by having them l
• astrovela/vaneA multimodal data engine built on a DuckDB fork that lets you process images, video, audio, and text and run L
• clayboxerreproduce/meccha-chameleon-toolAn external Windows tool that overlays ESP, a color picker, and a radar onto the hide and seek game MECCHA CHA
• goldyubrain/videosubtitlerA Windows tool that transcribes a video, translates it to Traditional Chinese, and burns the subtitles into th
• hiangx-robotics/metafineDiagnostic benchmark framework for robot manipulation policies that splits a task into separate understanding,
• mercuriustech/odyseus-spatial-vlmOdyseus Spatial VLM combines a vision language model with a depth estimation model to turn a single photo into
• nanovisionx/raev2Official PyTorch code for Improved Baselines with Representation Autoencoders, reaching state-of-the-art laten
• njsmith/pysrilmA minimal Python wrapper around SRILM, an older language modeling toolkit, letting you query word and sentence
• simplified-reasoning/su-01SU-01 is a 30B-parameter AI model trained to solve hard math and science olympiad problems with gold-medal-lev
• tencent-hunyuan/hunyuan-7bTencent's open-source AI language models (0.5B to 7B sizes) you can run on your own hardware. They support fas
• yt-dlp/yt-dlp-sample-pluginsA sample plugin package showing developers how to structure and install a Python plugin that extends yt-dlp, t
• 8bit64k/cronalyticsCronalytics is a Hermes Agent plugin that tracks the cost, token usage, and performance of scheduled AI agent
• alibaba-damo-academy/rynnworld-4dA research AI system that predicts color, depth, and motion together to help robot arms understand 3D scenes a
• allen-110/exploitariumA consolidated archive of one researcher's public proof-of-concept writeups covering vulnerabilities in tools
• sihengtao/phd-application-plannerA Claude Code/Codex skill that researches PhD programs and advisors, then builds an interactive dashboard to c
• wunderlandmedia/launchworthyA Claude Code skill that audits AI-built apps across five domains and produces a scored, copy-paste ready fix
• ycyao216/dexverseA research benchmark for testing robot hands on dexterous manipulation tasks, built on NVIDIA Isaac Sim.
• zostaff/ai-quant-researcherA Python framework where Claude proposes and backtests trading strategies against statistical rigor checks lik
• diabloidyobane/driverscopeA Python pipeline that hunts vulnerable Windows kernel drivers by scanning imports, extracting IOCTL surfaces,
• lingyichen-ai/openskillsOpenSkills lets you give AI agents modular skills—like summarizing meetings or processing PDFs—using simple ma
• nv-tlabs/artifixerAn NVIDIA research project that uses a diffusion model to fix visual errors and fill in missing camera views i
• ssrsec/webstrikeA browser extension based command and control framework for authorized red team engagements, letting a tester
• thehimel/async-document-processingAn async Python backend for processing documents through a web API, built with FastAPI and PostgreSQL.
• hamid-k/nginx-rift-private-labA private lab reproducing a critical NGINX heap overflow vulnerability, CVE-2026-42945, including an ASLR bypa
• juneyaooo/lineage-skillA pipeline that turns a video course's materials into an installable AI agent skill so an agent can answer que
• nvlabs/spatialclawSpatialClaw is a research framework from NVIDIA and KAIST where an AI agent writes step-by-step Python code to
• shootthesound/comfyui-meshA tool that splits large AI image and video models across two GPUs, using video compression hardware to speed
• afadtc/afa-dtc-skillsAn AI advisor system of 30 modules that gives independent e-commerce store owners strategic guidance on market
• alibaba-multimodal-industrial-ai/industrybenchA benchmark dataset and toolkit that tests how well AI language models understand specialized industrial and m
• alonera/aevumA free, ad-free desktop app that downloads video or audio from nearly any website using yt-dlp and FFmpeg unde
• mitkox/skilloptA tool that automatically rewrites an AI agent's system prompt by testing it, finding failures, and improving
• sanjayragavendra/mecchachameleon-autopainttoolA downloadable Windows cheat for the game Meccha Chameleon adding auto-camouflage, wallhacks, aimbot, and anti
• amyxvalen/flash-usdt-senderA command-line tool that generates crypto transactions designed to appear real for a limited time; its own rep
• echo-team-joy-future-academy-jd/echo-infinityA research project that generates continuous AI video for hours or even 24 hours straight by compressing video
• lifeiteng/omnivad-kitA lightweight toolkit that detects when someone is speaking in an audio file and can also spot singing or musi
• ppetrichor/worlddirectorWorldDirector generates controllable video simulations of a world, using an AI planning step plus a video mode
• shefyyuri/xai-cortexA Python library that manages conversation memory for AI chatbots, trimming or compressing older content so im
• stonedmodder/auto-tmcA desktop app that simplifies installing and managing TMC game modifications by automatically downloading, ver
• 1ove9/antenna-forgeAn AI platform that automatically designs radio antennas and verifies every design using real electromagnetic
• ali-vilab/diffusionopdResearch code for a new method that trains one AI image-generation model to excel at multiple goals at once by
• d2l-ai/d2l-ptAn open-source, interactive deep learning textbook in Portuguese that mixes explanations, math, diagrams, and

All repos · Explain any other repo →