This repository contains a monolithic Python CLI tool for downloading Udemy courses, including DRM-protected video content. The core logic in main.py orchestrates authentication (bearer token or browser cookies), fetches course curriculum via Udemy's private API, and downloads lectures, subtitles, assets, quizzes (normal and coding assignments), practice assignments, and role-play exercises. It relies heavily on external tools: yt-dlp for DASH/HLS segment downloading, aria2c for accelerated downloads, ffmpeg for muxing and video transcoding, and shaka-packager for DRM decryption. The code is functional but lacks modularity, with global state, sparse error handling, and a monolithic main script. The standout component is mp4parse.py, a well-structured MP4 box parser that extracts PSSH headers for Widevine key identification, which could be extracted and reused independently. The project is no longer actively maintained as stated in its README, and some dependencies may be outdated.
This is a personal interview preparation and learning repository focused on large language model (LLM) algorithm positions. It contains extensive interview question logs from 2024-2026, organized by company, covering topics such as fine-tuning, RAG, RLHF, agent design, model architecture, and deployment. The repository also includes learning notes on transformer, CNN, RNN, LoRA, PEFT, and other LLM-related concepts, along with code implementations: a decoder-only transformer trained on Chinese text (红楼梦), a reinforcement learning-based code optimizer using PPO, and various PyTorch examples. The utility scripts (DuckDB cache decorator, JSON extraction, logging, proxy management) are well-crafted and potentially reusable. However, the code components are explicitly unmaintained (the author states only interview records will receive future updates), and many paths are hardcoded. The interview notes are unstructured personal logs, not reusable as data assets without significant cleanup. The repo serves as a portfolio of technical capability and study material rather than a production system or reproducible data pipeline.
This repository is a curated index and automation hub for backing up visually impressive websites. The core code is a single Node.js script (`autoreadme/index.js`) that runs via GitHub Actions on a daily cron schedule. It counts screenshot files in the `screenshot/` directory, updates the README badge with the total count, randomly selects a screenshot to feature in the 'Daily Show' section, and posts a tweet with that image and a link to the corresponding backup repository. The actual website backups are stored in separate repositories (Pack01–Pack06), while this repo serves as a catalog with screenshots and links. The automation script demonstrates OAuth 1.0a integration with the Twitter API, file I/O, and regex-based README manipulation. The repository is not a software application in the traditional sense; it is a maintenance tool for a curated list. The code is functional but tightly coupled to this specific project's structure, limiting reusability.
This repository aggregates and periodically syncs M3U playlists for Chinese live-streaming platforms (YY, Bilibili, Douyu, Huya) from a third-party proxy service (live.ottiptv.cc). The sole automation is a GitHub Actions workflow that downloads the playlists every 2 hours and commits any changes. There is no application code, data transformation, or analysis; the provided files are directly consumable by IPTV players. The repository includes placeholder test streams and many duplicate entries, indicating mechanical collection. Legal disclaimers in the README warn against commercial use and access from China, suggesting copyright risks. The content is entirely dependent on the external proxy service and platform availability, which introduces high fragility. The last commit date is implausibly far in the future (2026), likely a placeholder, implying the repository is not actively maintained despite the workflow definition.
This repository is the companion codebase for the Cyfrin Updraft 'Smart Contract Auditing, Assembly, Security, and DeFi Ultimate Course'. It contains no running application or service; instead, it provides educational materials: markdown guides for how to ask questions, answer questions, structure audit findings, and onboard for security reviews, along with issue templates and a single Solidity example file (eip712hashing.sol) demonstrating EIP-712 hashing. The repository also hosts images and links to external course videos. There are no executable scripts, API endpoints, or data pipelines. The primary function is to serve as a reference and resource for students following the video course.
This repository is a curated 'awesome list' of job boards, organized by categories such as general, AI, data, blockchain, design, programming, remote, and by country. It contains no executable code, data pipelines, or analysis; it is purely a static markdown document with links and brief descriptions. The primary function is to serve as a reference directory for job seekers, not as a software application or data research project. Despite its popularity (1000+ stars), there are no reusable components, tests, or engineering patterns. The list is actively maintained with frequent additions, but the content is entirely editorial.
Claude Delegator is a Claude Code plugin that provides GPT (via Codex CLI) and Gemini (via Gemini CLI) as specialized expert subagents through MCP (Model Context Protocol). It includes five expert prompts—Architect, Plan Reviewer, Scope Analyst, Code Reviewer, and Security Analyst—along with orchestration rules that instruct Claude Code when and how to delegate tasks to these external AI experts. The architecture is orchestrator-based: Claude Code acts as the central orchestrator, detecting delegation triggers, selecting the appropriate expert, building a structured prompt, calling the external AI via MCP, and synthesizing the response. The plugin supports both single-shot and multi-turn conversations, with context preserved via threadId for chained tasks. The codebase consists primarily of markdown configuration files (prompts, rules, commands) and a single JavaScript file implementing a custom MCP bridge server for Gemini CLI, since Gemini lacks native MCP support. The Gemini bridge (`server/gemini/index.js`) is a well-crafted, zero-dependency JSON-RPC 2.0 server that wraps the Gemini CLI with robust input validation, error handling, and protocol compliance. The prompts and rules are adapted from the oh-my-opencode project and are well-structured but not original. The project is now deprecated and no longer maintained. The Codex CLI removed the `codex mcp-server` command that this plugin relied on, breaking GPT delegation entirely. The Gemini bridge is untested against current Gemini CLI releases. The repository serves primarily as a reference implementation for MCP-based AI orchestration patterns, and the Gemini bridge code remains a reusable component for wrapping any CLI as an MCP server.
This repository is the source code for the CS 6120 (Advanced Compilers) course website at Cornell University. It is built using the Zola static site generator and contains lesson plans, a course blog, a schedule, a syllabus, and supporting materials in Markdown, TOML, and HTML templates. The repository is not a reusable software application or library; it is a content repository for a specific course. The codebase includes Zola configuration, Sass for styling, and a GitHub Actions workflow for building and deploying the site. The 'application' label is used because the site is a web application, but the emphasis is on content delivery rather than software engineering craftsmanship.
This repository provides unofficial precompiled binary wheels for a comprehensive set of geospatial Python libraries (e.g., GDAL, Fiona, Shapely, rasterio) on Windows, including Windows ARM64. The repository itself contains no source code; it functions purely as a distribution channel and release archive. The wheel builds incorporate numerous statically or dynamically linked native libraries (GEOS, PROJ, HDF5, etc.) and target recent Python versions (3.12–3.15). Build environment details are documented in the README (Visual Studio 2022, Windows Dev Kit, Rust). This is not a reusable software project but a convenient binary distribution service for the geospatial Python ecosystem on Windows.
This repository is a static resource list providing curated links to academics on Mastodon. It organizes resources by discipline, institution, groups, servers, and other categories, all presented in a single README.md file. The project does not contain any code, data processing logic, or interactive components; it serves as a reference directory for academics seeking to engage with the Mastodon federated social network. The repository is maintained, with the most recent commit indicating recent updates to the resource list. There is no technical drift as the project contains no executable code or dependencies that could become outdated. While the resource list is useful, it lacks reusable software components or novel engineering patterns, representing a straightforward compilation of externally hosted resources.
This repository is a companion to the Packt book 'Python Feature Engineering Cookbook' by Soledad Galli. It contains over 70 Jupyter notebooks organized by chapter, each demonstrating a specific feature engineering technique for machine learning, including missing data imputation, categorical encoding, variable transformation, discretization, outlier handling, date/time features, scaling, feature creation, and text feature extraction. The notebooks use public datasets (Titanic, Boston Housing, Credit Approval UCI, 20 Newsgroups) and implement techniques both manually with pandas and using specialized libraries like feature_engine, category_encoders, and featuretools. While educational and self-contained, the code is not production-grade and relies on outdated library versions and deprecated datasets (e.g., load_boston), making it a tutorial resource rather than a reusable software project. The repository is stale—last updated around 2020—with no ongoing maintenance, and missing dependencies (e.g., category_encoders not listed in requirements.txt) further reduce reproducibility.
This repository is a curated list of command-line tools, designed to help users discover and evaluate CLI utilities for various tasks. It does not contain any executable code or application logic; instead, it serves as a static reference document hosted on GitHub. The main content resides in `readme.md`, which organizes tools into categories such as CSV, File System, Git, Networking, and Programming Languages, each with links to the tools' source repositories and concise descriptions. Contribution guidelines are provided in `contributing.md` to maintain the list's quality and consistency. The repository uses shields.io badges to display metadata like the license and links to related curated lists. Although the repository lacks reusable software components or architectural patterns due to its static nature, it is well-maintained with recent commits and active issue tracking. It exhibits low technical drift as it has no code dependencies to become outdated. As a reference resource, it effectively fulfills its goal of aggregating CLI tools, though it does not demonstrate notable software engineering craftsmanship.
This repository hosts the official Sinatra website and documentation, published at sinatrarb.com. It is a static site built with Jekyll, using Ruby Sass (Bourbon/Neat) for CSS and Gulp for asset processing. The site aggregates comprehensive documentation for the Sinatra web framework, including the core README, Sinatra::Contrib extensions, Rack::Protection middleware, release notes, FAQ, and a community showcase. The build pipeline (Rakefile) automatically pulls latest documentation from the sinatra, sinatra-contrib, and rack-protection upstream repositories. The repository contains no running application code; it is purely a content and documentation site. The build tooling is significantly outdated, relying on deprecated Ruby Sass and old versions of Bourbon/Neat, and there are signs of domain and configuration drift (e.g., references to sinatra.github.com). The last commit date (2026) appears to be anomalous, and the repository shows no active maintenance.
This repository is a Chinese-language content mod for the Twine/SugarCube game 'Degrees of Lewdity' (DoL). It adds a large expansion including a City Library location, new NPCs (Seath, Alice, Liddell, Zephyr), new weapons, school mechanics (prefect system, free-listening permission), a gym, bicycle transport, live-streaming minigame, parasite/tentacle combat, and book-based traits. The mod is built using DoL ModLoader's patching system (TweeReplacer, ReplacePatcher, DoLTimeWrapperAddon) and injects content into base game passages and JavaScript. The codebase is functional but exhibits significant technical debt: large amounts of commented-out code, disused folders, informal developer comments expressing frustration, incomplete features, and fragile patches targeting specific base-game strings. The mod is primarily maintained by a single developer who acknowledges infrequent GitHub checks. Despite its messiness, the mod demonstrates non-trivial integration with DoL's event system and provides a substantial amount of new gameplay content.
ElysiaOS is a personal Linux desktop dotfiles and ricing repository, not a distributable operating system, an application server, or a data project. The repository contains a complete themed Arch Linux + Hyprland setup: window manager configs, Waybar status bars, EWW dashboard widgets, swaync notifications, rofi launchers, kitty terminal configs, SDDM login theme, Plymouth boot splash, icon/cursor/wallpaper assets, and an installer script that wires everything together. The dominant character is a heavily customized desktop environment tied to a single author's machine and anime-themed aesthetic.
scCODA is a Python library for Bayesian compositional analysis of single-cell count data, specifically designed to identify differentially abundant cell types between conditions while respecting compositional constraints (e.g., cell type proportions sum to 1). The core implements a Dirichlet-Multinomial model with spike-and-slab priors, providing posterior inclusion probabilities and effect size estimates. It integrates with the scverse ecosystem (anndata, scanpy) for data handling and uses TensorFlow Probability for HMC/NUTS sampling. The package also includes wrappers for alternative methods (ALDEx2, ANCOM, etc.) and provides utilities for data conversion, visualization, and synthetic data generation. While the repository is well-architected with clear separation of concerns and includes tutorials and unit tests, it is explicitly marked as no longer maintained (successor in pertpy using JAX). The technical drift is high due to outdated TensorFlow/TPP dependencies and lack of active development.
MetaXcan is a command-line and library framework for transcriptome-wide association studies (TWAS). It integrates GWAS summary statistics or individual-level genotype data with gene expression prediction models to compute gene-level association statistics. The core methods implemented are S-PrediXcan (summary-based), PrediXcan (individual-level), MultiXcan, and S-MultiXcan (multi-tissue meta-analysis). The architecture consists of Python scripts that orchestrate data processing steps (e.g., parsing GWAS, aligning alleles, computing z-scores) using a core library (`metax`) with modules for genotype reading, GWAS loading, prediction model management, and covariance matrix handling. The system is designed for bioinformatics researchers and requires external reference data (e.g., GTEx models, 1000 Genomes LD). While functional, the codebase exhibits significant technical debt: Python 2 compatibility remnants, deprecated modules, hardcoded URLs, and limited test coverage. The GUI component (Tkinter) is unmaintained. The repository is hybrid, containing both application scripts and data processing pipelines, but the dominant character is a software tool for running TWAS analyses.
sysmon is a system monitoring and task management application that provides real-time visualization of system metrics through a web interface. It consists of a backend driver service built with Python/Falcon that exposes API endpoints for system data, and a frontend service available in both Python/Flask and Go implementations that serves an interactive dashboard. The application collects CPU, memory, disk, network, sensor, and process information using the psutil library, displays it via real-time charts (using SmoothieCharts) and tables, and allows process management operations (terminate, kill, suspend, resume) through a passcode-protected interface. The architecture features a clean separation between data gathering (hard.py), API endpoints (falc.py), and presentation layer (frontend templates and static assets).
Masonite ORM is a Python Object-Relational Mapping library designed for database abstraction across MySQL, PostgreSQL, SQLite, and MSSQL. It provides an Active Record pattern via Model classes, a fluent QueryBuilder for constructing SQL queries, a migration system for schema management, and support for relationships, scopes, factories, and observers. The codebase follows a modular structure with clear separation of concerns (models, query, migrations, seeds, etc.) and includes a CLI for common database operations. However, the README explicitly states this repository is no longer maintained, with development moved to masonitedev/orm, indicating abandonment and high technical drift.
SecLists is a massive, curated collection of security testing wordlists and payloads, not a software application. It serves as a companion dataset for penetration testers and security researchers, aggregating lists of usernames, passwords, URLs, fuzzing payloads, web shells, and other sensitive data patterns into a single repository. The 'architecture' is purely organizational: a directory tree of text files grouped by category (e.g., Passwords, Fuzzing, Discovery, Payloads), supplemented by a set of maintenance scripts in `.bin/` for updating lists from remote sources, validating file integrity, and automating contributions via GitHub Actions. The primary 'data flow' is curation: lists are sourced from external projects (e.g., FuzzDB, RAFT, Probable-Wordlists, sqlmap), committed to the repo, and periodically refreshed by automated scripts. There are no models, analytics, or dashboards produced. The repository's value lies entirely in the breadth and quality of its aggregated lists. Maintenance is active, with commits as recent as June 2026 and automated hourly/daily update workflows for select lists (JWT secrets, API endpoints, etc.). However, from a data engineering perspective, the repo has significant limitations. There is no dependency management (no requirements.txt, Dockerfile, or environment spec), no data validation beyond basic newline/empty-line checkers, and no documentation of data lineage or update frequency for the vast majority of lists. Many READMEs are outdated or reference defunct sources. The collection is highly heterogeneous: some lists are meticulously curated and documented, while others are raw dumps with no metadata. There are no tests, no schema definitions, and no quality metrics. The `.bin/` scripts are functional but minimal—straightforward Python and shell scripts for fetching remote files and checking formatting, none of which demonstrate non-obvious engineering craftsmanship. In summary, SecLists is an invaluable reference dataset for the security community, but it is not a data pipeline or analytical project. It is a file repository with light automation, judged by data engineering standards: low reproducibility, minimal documentation, no dependency management, and no data quality framework beyond basic file hygiene.