FishNet is a comprehensive Unity multiplayer networking framework providing a full client-server architecture with RPCs, sync types, prediction, scene management, observer conditions, and multiple transport backends (LiteNetLib, Synapse, Yak). It employs IL code generation via Mono.Cecil to auto-generate serializers and RPC stubs, achieving high performance and low bandwidth usage. The codebase is well-structured with extensive use of partial classes, pools, and utility patterns. It includes demos for authentication, prediction, collider rollback, and scene management. While the framework is tightly coupled to Unity, many utility components (e.g., object pooling, ring buffers, quaternion compression) are reusable in isolation and demonstrate strong engineering craftsmanship. The project is actively maintained with a large community and regular commits.
MonitorControl is a macOS menu-bar utility that controls external displays' brightness, contrast, and volume. It does not just send one kind of command; it uses multiple control paths: hardware DDC/CI over IOKit I2C on Intel Macs, DDC/CI over IOAVService on Apple Silicon, Apple's private DisplayServices API for Apple/Thunderbolt displays, gamma-ramp adjustments for software dimming, and transparent shade windows for virtual displays. User input flows from native media keys, custom keyboard shortcuts, and menu-bar sliders into a DisplayManager singleton, which coordinates native OSD overlays, smooth brightness transitions, and cross-display synchronization. The app uses Sparkle for updates, SimplyCoreAudio for audio device handling, MediaKeyTap for media-key interception, and KeyboardShortcuts for user-defined bindings. Architecturally, this is a classic Cocoa application with a storyboard-based preferences UI, a login helper app (MonitorControlHelper), and a clean separation between models (Display, AppleDisplay, OtherDisplay), support services (DisplayManager, DDC backends, OSD utilities, media-key handling), and view controllers for preferences and onboarding. Entry points are main.swift, AppDelegate, DisplayManager, and MenuHandler. Persistence is UserDefaults backed through a PrefKey enum, and the UI is localized into 19 languages. There is no data/research component; this is purely an end-user utility. The codebase is modular and well organized, with the DDC communication isolated into Intel and ARM backends and display-specific logic abstracted behind a base Display class. It is actively maintained, with recent commits and a modern dependency setup. Technical drift is medium rather than low because much of the functionality depends on private or undocumented macOS APIs (DisplayServices, CoreDisplay, OSDUIHelper) and the code contains many macOS-version-specific workarounds, but the project tracks current macOS releases and continues to update its fallbacks.
Optio is a self-hosted, Kubernetes-native orchestration platform for AI coding agents (Claude Code, OpenAI Codex, GitHub Copilot, Gemini, Cursor, OpenCode). In production it runs three task tiers — repo tasks (issue/request → agent pod → PR), standalone jobs (parameterized agent runs without repo checkout), and persistent agents (long-lived, message-driven agents with turn-based execution) — plus an autonomous PR review loop. The core loop is driven by ~15 BullMQ workers (task, PR watcher, review, webhook, ticket sync, reconcile, workflow trigger) against a Postgres database, with Redis pub/sub powering a real-time event bus and WebSocket log/terminal/chat streaming to the Next.js dashboard and CLI. Agents execute in ephemeral pods built from per-language Docker images; a K8s-style reconciliation control plane (world snapshot + pure decision functions + CAS-gated executor) converges four run kinds and catches drift. The architecture is a pnpm/turbo monorepo: a Fastify 5 API with Drizzle ORM and Zod-validated OpenAPI schemas; a Next.js 15 admin dashboard (tasks, jobs, reviews, sessions, agents, cluster, analytics); a Commander-based CLI (login, task/repo/session/secret commands over REST + WebSocket); and a static docs site. Shared packages provide a clean AgentAdapter interface (`validateSecrets` → `buildContainerConfig` → `parseResult`), a Docker/Kubernetes container runtime abstraction, ticket provider adapters (GitHub, GitLab, Jira, Linear, Notion), and shared types/utilities (task state machines, repo URL parsing, SSRF guards, cost arithmetic). Workspace-scoped multi-tenancy with admin/member/viewer RBAC, AES-256-GCM secret encryption with AAD binding, WebSocket upgrade tokens, HMAC-signed internal calls, and a security-focused audit trail (optio_actions) are enforced throughout. Engineering strengths: the test suite is unusually thorough for a project this size — nearly every route, service, worker, and shared utility has a unit test, with per-file Postgres/Redis isolation for integration tests, real-server E2E tests using a deterministic fake container runtime (with mock directives like [[mock:pr]]), and Playwright smoke tests. The reconcile pattern, generic state machine utilities, and pure decision functions in `packages/shared/src/reconcile/` are the highest-craftsmanship code. Known debt: five near-duplicate NDJSON event parsers (one per agent, with drift risk), hardcoded model pricing tables in the Codex/Gemini adapters, scattered duplicated workspace-scoping checks, a 60+ file migration history with historical prefix collisions (now script-enforced), and raw SQL in the analytics service. The project moves fast (v0.1 → v0.4.1 within months, last commit 2026-08) and some surface area (OpenCode/OpenClaw adapters, skills marketplace) is explicitly experimental, but the core architecture is coherent and production-oriented.
This repository provides a generic-based dependency injection framework for Go, called 'kod' (Killer Of Dependency), along with a CLI tool for code generation, component registration, and dependency graph validation. It features a rich interceptor chain for cross-cutting concerns (retry, validation, tracing, rate limiting, circuit breakers, etc.), and includes a comprehensive test suite and examples. The codebase also incorporates an extensive set of AI agent skills for software development workflows (grilling, code review, TDD, domain modeling) from the mattpocock/skills ecosystem. The primary value is the DI framework that leverages Go generics to provide compile-time safety for component wiring, with generated boilerplate code. The architecture is modular with clear separation of core, interceptors, internal utilities, and tests. It is actively maintained with modern Go features.
Tax-Calculator is a mature, open-source microsimulation library for static (and simple behavioral) analysis of the US federal individual income and payroll tax system. It provides a Python API and a CLI tool (`tc`) that allow users to define policy reforms via JSON parameter files, apply them to microdata samples (CPS, PUF, or TMD), and produce distributional tables, revenue estimates, marginal tax rates, and welfare measures. The core architecture separates policy parameters (`Policy`), taxpayer records (`Records`), and economic adjustments (`GrowFactors`, `GrowDiff`, `Consumption`) from the calculation engine (`Calculator`), which uses Numba JIT compilation for performance. While primarily an application/library with formal releases and CI/CD, it also bundles research data (CPS microdata, grow factors) and includes extensive cross-validation against NBER's TAXSIM-35 model. The codebase is well-structured, heavily tested, and actively maintained, with a clear modular design that supports both CLI and programmatic use.
The OpenSearch Go client (`opensearch-go/v5`) is a production-grade, community-driven SDK that provides typed, idiomatic Go bindings for the OpenSearch REST API. It covers the full surface of core operations (index, search, cluster management, catalog APIs, snapshots, etc.) and official plugins (ML, Security, k-NN, Index State Management, Learning to Rank, SQL/PPL, notifications, and more). The client is not just a thin HTTP wrapper—it includes a sophisticated transport layer with adaptive connection pooling, node discovery, retry policies, shard-aware routing, and pluggable observability (Prometheus, OpenTelemetry) via an observer pattern. Additionally, it ships a code generator (`cmd/osgen`) that produces all typed request/response structs and tests from the OpenSearch OpenAPI specification, and a migration linter (`cmd/osapilint`) to automate major-version upgrades (v2→v5). Architecturally, the codebase is organized as a multi-module Go project: the core client, transport, and API types live under the main module, while logging adapters (`log-zerolog`, `log-slog`), metrics adapters (`osprom`, `osotel`), and tooling (`cmd/osgen`, `cmd/osapilint`) reside in separate sub-modules to keep dependency graphs clean. The primary entry points are `opensearch.NewClient()` / `opensearchapi.NewClient()` for the core API, and plugin-specific constructors (e.g., `ml.NewClient(osClient)`) for extended capabilities. Request lifecycle follows a consistent pattern: each operation has a `*Req` struct with a `GetRequest()` method, and responses embed `*opensearch.Response` with `Inspect()` and `RawBody()` accessors. Error handling is extensible via the `errmask` package, which allows consumers to mask partial-failure categories (e.g., bulk item errors) from surfacing as Go errors. Data flows through the client’s layered pipeline: user constructs a typed request → internal `build.Request()` and `params.Encode()` produce an `*http.Request` → the transport applies policies (connection selection, retry, node discovery) → performs the HTTP call → the response is deserialized into a typed struct. The transport uses garbage-avoiding patterns (sync.Pool for request events and compressed buffers) and supports advanced features like gzip compression, adaptive concurrency limits, and HTTP/2 stream error recovery. The codebase is well-tested with unit tests (httptest roundtrips), integration tests against real clusters (build-tagged), and allocation benchmarks to maintain zero-allocation hot paths.
RTXPT is a high-fidelity real-time path tracing reference implementation developed by NVIDIA. It serves as both a standalone demo application and a reference for integrating advanced rendering techniques into production engines. The codebase implements state-of-the-art path tracing with ReSTIR (Reservoir Spatiotemporal Importance Resampling) for direct and global illumination, NEE-AT (Next Event Estimation - Adaptive Temporal) for lighting feedback, and Stable Planes for denoising. It integrates heavily with NVIDIA RTX SDKs including RTXDI, NRD (Denoiser), OMM (Opacity Micromaps), and DLSS (Streamline). The application supports both DirectX 12 and Vulkan backends via the nvrhi abstraction layer, though Vulkan support is noted as work-in-progress. The architecture follows a modular design with separate components for lighting bakers, material systems, shader compilation, and rendering passes. It uses the Donut engine framework for scene management and shader compilation infrastructure.
aliyundrive-webdav is a WebDAV server application that bridges WebDAV clients (like Infuse, nPlayer) with AliyunDrive (Alibaba Cloud Drive) storage. The system is built in Rust and uses the dav-server crate as its WebDAV foundation. Key functionality includes: authenticating with AliyunDrive via refresh tokens (obtained through command line, environment variables, or QR code login), translating WebDAV operations (PROPFIND, GET, PUT, etc.) to AliyunDrive API calls, implementing caching for directory listings to reduce API calls, handling file uploads via chunked resumable uploads, and providing automatic token refresh in the background. The architecture consists of modular components: cache (directory entry caching), drive (AliyunDrive API communication), login (QR code authentication flow), vfs (virtual file system layer), and webdav (server implementation and auth). External dependencies include the AliyunDrive API, reqwest, tokio, tracing, clap, moka, dashmap, serde, and optional TLS dependencies. The application is actively maintained with frequent commits, CI/CD pipelines for multi-platform builds, and support for Docker, Snap, OpenWrt, and systemd deployments.
This repository is the `eslint-plugin-react` package, a widely-used ESLint plugin that provides over 100 lint rules for React and JSX codebases. It is not a standalone application but a library consumed by ESLint to enforce best practices, prevent common errors, and maintain code style in React projects. The plugin supports both legacy `.eslintrc` and modern flat config (`eslint.config.js`) formats, and works with multiple parsers including babel-eslint and TypeScript-ESLint. The architecture is modular: each rule is a self-contained module in `lib/rules/`, sharing common utilities from `lib/util/` for component detection, AST traversal, version checking, and error reporting. The codebase is well-tested with a comprehensive test suite that runs across parser versions and ESLint versions. It is actively maintained with frequent releases and community contributions.
This repository is a full-stack Next.js application that serves as the world's largest open-source prompt library for AI. It provides a platform for users to create, share, discover, and manage prompts for various AI models (ChatGPT, Claude, Gemini, etc.). The core functionality includes CRUD operations for prompts with types (text, image, video, audio, skill, taste), voting, commenting, versioning, change requests, collections, and moderation tools. It features AI-powered capabilities such as semantic search via OpenAI embeddings, prompt improvement, translation, quality checking, and media generation through plugin-based providers (Fal.ai, Wiro.ai). The application supports multi-tenancy for self-hosting with configurable branding, authentication plugins (OAuth, OIDC, GitHub, Apple, Google), and storage backends (S3, DO Spaces). Beyond the web app, it includes a CLI tool (packages/prompts.chat) for browsing and running prompts locally, a Raycast extension for quick access, and an MCP server for integration with AI coding assistants. Additionally, it bundles an interactive educational book on prompt engineering (28+ chapters) and a gamified kids learning section, both fully localized into 17 languages. The codebase is well-structured with a plugin registry pattern, configuration-driven design, and extensive use of modern Next.js patterns (App Router, server components, streaming).
MsgVault is a self-hosted, local-first personal communications archive system. It ingress messages from email (IMAP, Gmail, mbox, PST), chat (Slic, Discord, Matrix, Beeper, WhatsApp, Facebook Messenger), SMS, meeting ttranscripts (Granola, Circlback, Plaud, Muelsi), and contacts (CardDAV, vCard). It store them in SQLite or PostgreSQL with full-text and vector embeddings search, provides a rich HTTP API, a Svelte web UI, a CLI, and an MCP serve for AI tools. The system es also feature person identity matching across source, contact syn, document indexing, and telemetry. Technically, it is a monolithic Go binary with modular packages, abdecking both interactive use and automated daemon operation. While it is alpha software with frequent API changes, its architecture is well-structure, extensively tested, and includes several reusable component like robust error classification, HTTP retry logic, and cross-process cache locking. The frontend uses generated TypeScript clients from the OpenAPI spec, and the entire system is deployable via Docker or native binaries.
The PyDev.Debugger (pydevd) is a Python debugger backend designed for IDE integration and remote debugging. It provides core debugging capabilities including breakpoints, stepping, variable inspection, and expression evaluation. The architecture features a client/server model for remote debugging, with in-process tracing via settrace(). Key entry points include pydevd.py (main debugger), pydevd_attach_to_process (live process attachment), and pydevconsole.py (interactive console). The debugger supports multiple protocols (legacy line-based and JSON/DAP for VS Code integration) and employs Cython optimizations with fallbacks to pure Python. It includes extensive test suites covering edge cases like multiprocessing, async, GUI event loops, and framework-specific integrations (Django, Flask). The code demonstrates strong modularity through extension points for custom variable rendering and plugin architecture.
Creepyface is a JavaScript library that makes HTML images follow the pointer (mouse/finger/touch) by switching between different angled images. It consists of a core library, React and custom element wrappers, and point provider plugins for dance, firefly, and tilt tracking. The repository also includes a Next.js website that demonstrates the library and allows users to create and share their own creepyfaces. The architecture uses a monorepo with Lerna/Yarn workspaces, featuring a clean separation between the core library logic, UI wrappers, and the demonstration site with its own backend for image processing and storage.
torlink is a terminal-based torrent client that combines a rich Text User Interface (TUI) with headless daemon modes for torrent search, download, seeding, and file serving. Built with TypeScript, Node.js, and React (via Ink for TUI rendering), it provides zero-configuration access to curated torrent sources across categories like games, movies, TV, and anime. The application features parallel source querying, result deduplication, automatic seeding with optional file deletion, and persistent state management for queues, history, and seeds. The architecture follows a modular design: the TUI layer (src/ui) handles interactive modes, daemon services (src/daemon) manage headless operations (serve, files, watch, seed, attach), core download logic (src/download) manages the torrent queue and engine integration, and source plugins (src/sources) fetch torrents via RSS, APIs, or web scraping. Key utilities include atomic file writes, resilient HTTP fetching, caching, and format helpers. The CLI entry point (scripts/cli-entry.cjs) handles Node version checks and WebRTC fallbacks before launching the appropriate mode. External dependencies include Ink for TUI rendering, webtorrent for torrent handling, node-datachannel for WebRTC fallbacks, and various HTTP clients. The project maintains active development with recent commits, comprehensive test coverage, and CI workflows. While dependency pinning indicates managed compatibility risks, the codebase demonstrates strong engineering practices through isolated utilities, reactive patterns for concurrent searches, and cross-platform path handling.
Kronotop is a distributed database system built on FoundationDB that provides Redis-compatible (RESP2/RESP3) interfaces for multiple data models including key-value (stash), document (bucket with BSON storage and querying), and sorted sets (zmap). It features ACID transactions, secondary indexing (including vector search via HNSW), clustering, sharding, replication, and namespace isolation for multi-tenancy. The architecture is modular with clear separation between network handling (Netty), protocol processing, service layers (transaction, session, data services), and storage (volume-based segment store). Key entry points include the server entry point (KronotopInstanceStarter), CLI (kronotop-cli), and control tool (kronotop-ctl). The system relies on external dependencies such as FoundationDB, Netty, Jackson, Lettuce, Guava, JVector, and Typesafe Config. The codebase demonstrates strong engineering practices with reusable components like versionstamp utilities, retry mechanisms, and protocol handling.
The TodoMVC repository serves as a collection of reference implementations of the TodoMVC specification across numerous frontend frameworks. It provides a static file server (server.js) to serve these examples and includes comprehensive Cypress-based test infrastructure to validate implementations against the specification. Each framework implementation is self-contained in the examples/ directory, with shared assets in todomvc-common/ and todomvc-app-css/. The repository enables comparative evaluation of frameworks through identical feature implementations and automated testing.
ServerStatus is a lightweight server monitoring and cloud dashboard system. It consists of a Go-based server that accepts TCP connections from Python clients, collects system metrics (CPU, memory, disk, network, processes, ping latency), and exposes a real-time JSON stats endpoint and a full REST management API. The server also supports service monitoring (HTTP/TCP probes), SSL certificate expiry checks, and a configurable watchdog rule engine that evaluates expressions against live metrics and triggers HTTP callbacks. The web UI is a single-page application with filtering, sorting, detail drawers, and an admin panel for CRUD operations on configuration. The architecture is a classic client-server model: multiple Python agents (client-linux.py or client-psutil.py) connect to a single Go server over a custom TCP protocol. The server authenticates agents, receives periodic metric updates, and serves a web dashboard and API. The Go server is a monolith with separate modules for HTTP handling, TCP agent handling, configuration management, SSL checking, and watchdog evaluation. It uses the Gin web framework for HTTP and the expr-lang library for rule evaluation. Key entry points are the TCP agent listener (port 35601), the HTTP server (port 80/8080), and the CLI flags for configuration. External dependencies include Go modules (gin, expr-lang), Python libraries (psutil for the psutil client), and Docker for containerized deployment. The system is designed for easy deployment via Docker Compose or shell scripts, with environment variable overrides for all client and server settings.
This repository provides a two-skill workflow system for academic paper creation designed to work with Claude Code. The 'academic-paper-strategist' skill handles research planning through platform analysis, literature search, gap identification, and outline optimization with quality gates. The 'academic-paper-composer' skill executes systematic writing from outlines with chapter-level quality checks and final evaluation. The system emphasizes quality control through structured rubrics and iterative validation, transforming research ideas into submission-ready manuscripts with comprehensive quality reports. While primarily documentation-heavy, it contains several self-contained, reusable evaluation systems that demonstrate sophisticated assessment frameworks.
RMQTT is a production-grade, high-performance MQTT broker written in Rust, supporting MQTT v3.1, v3.1.1, and v5.0. It is deployed as a standalone binary (`rmqttd`) or as an embeddable library, and features a modular plugin architecture with 26+ built-in plugins covering authentication (HTTP, JWT), authorization (ACL), storage (RAM, Sled, Redis, Redb), bridging to external messaging systems (Kafka, Pulsar, NATS, ReductStore, remote MQTT), clustering via Raft or broadcast, and a rich HTTP API with an embedded Vue.js dashboard. The broker handles millions of concurrent connections, supports TLS, WebSocket, and QUIC transports, and includes advanced features like shared subscriptions, delayed publishing, flapping detection, topic rewriting, and system topic subscriptions for observability. The architecture is based on the Tokio async runtime, with per-connection tasks, bounded message queues, and a plugin hook system that allows extending the broker without modifying core code. Key entry points include `rmqtt-bin/src/server.rs` for binary startup, `rmqtt/src/v5.rs` and `rmqtt/src/v3.rs` for protocol handling, and `rmqtt/src/context.rs` for runtime orchestration. External dependencies include gRPC (via Tonic) for inter-node cluster communication, Sled/Redis for persistent storage, and various client libraries for bridges (rdkafka, pulsar, nats). The broker is extensively tested with a custom test harness (`rmqtt-test`) that includes >600 functional, stress, and chaos tests, many reproducing real GitHub issues. The codebase demonstrates strong engineering practices: a clean plugin lifecycle, efficient topic trie for wildcard matching, reusable utility crates (`rmqtt-utils`, `rmqtt-codec`, `rmqtt-net`), and a comprehensive test framework that manages broker processes and configs. The dashboard and HTTP API provide operational tooling for client management, metrics, and plugin configuration. While the project is actively maintained, the large number of feature flags and optional storage backends create some configuration complexity, and a few tests rely on hardcoded ports or timing, which can be brittle. Overall, RMQTT is a solid, extensible MQTT broker suitable for IoT and real-time messaging use cases.
InvoicePrinter is a Ruby library for generating PDF invoices, with additional CLI and HTTP server interfaces. It takes structured invoice data (as Ruby objects or JSON) and generates formatted PDF documents using the Prawn library. The system consists of three main components: the core library for PDF generation, a command-line executable for direct use, and a server executable providing a HTTP API. The architecture shows good separation of concerns with distinct modules for data models, PDF rendering, font handling, and server interface.