2,454 tools and products — trending open source, and what gets used in AI and other work.
KubeRay is a platform for running Ray workloads within Kubernetes clusters. In the described DatologyAI pipeline, it was used with Ray and vLLM to orchestrate large-scale synthetic-data jobs on EKS on HyperPod.
Spark is a data-curation tool used to run jobs that prepare prompts and documents for synthetic-data generation. In DatologyAI's workflow, it supports the curation and synthesis stages used to produce large-scale training data.
EKS on HyperPod is an AWS infrastructure service for running Kubernetes-orchestrated H100 GPU clusters. DatologyAI used it as the execution environment for a large-scale synthetic-data pipeline combining Ray, KubeRay, and vLLM, with workflows spanning data curation, synthesis, training, and evaluation.
DatologyAI is a data curation service for AI model builders. It processes proprietary enterprise data, public web data, and licensed data for pre-training and mid-training on open models or privately deployed foundation models. Its four-stage pipeline cleans poorly formatted, empty, short, or evaluation-contaminated data; curates it by quality, task, and taxonomy; creates synthetic data for task relevance and diversity; and composes sources into staged training datasets. The company also describes seeded rephrasing methods for generating synthetic data at trillion-token scale.
Traversal is an AI site reliability engineering platform for complex enterprise production systems. It analyzes production data to triage alerts, identify incident root causes through causal analysis, and support incident resolution, with capabilities described for alert intelligence, production support, code resilience, and eventual self-healing fixes. Its AI SRE system is presented as operating at petabyte scale, with Traversal Workers described as proactive agents that can act without being prompted. The platform is listed as trusted by American Express, PepsiCo, and DigitalOcean.
Elastic is an enterprise search and data platform whose Observability offering uses machine learning and analytics to provide unified visibility into systems, including logs and metrics. In the cited discussion, it is described as helping identify broken or correlated systems, while stopping short of establishing the causal root of an incident. Elastic also provides search, security analytics, and agent-building capabilities through the Elasticsearch Platform.
Splunk is a data platform for security and observability. It helps organizations analyze operational data to identify broken or correlated systems and prevent downtime, although observability alone does not necessarily determine the root cause of an incident. The platform is designed to unify security and observability data at large scale and support AI-related workflows.
ServiceNow is an observability and incident-management platform used to manage operational incidents. Incident updates can be posted to ServiceNow tickets.
TidyCal is a calendar management and appointment-booking service for creating shareable booking pages, setting availability, and synchronizing calendars. It supports one-time and recurring bookings, session packages and bundles, automated meeting URLs, reminders, and payments through Stripe or PayPal. Clients can select a time and pay when booking, while providers can offer consultations, coaching, classes, or other services.
An MCP connection that lets Claude operate an Omnisend account through natural-language prompts. The video describes using it to add contacts, build segments, create and send email campaigns, and review campaign results.
Datadog Agent Observability is an AI-agent observability product for evaluating, improving, and tracing agents across development and production. It combines offline experimentation with production observability: teams can evaluate agents in production, build golden datasets from annotated traces, and test changes against those datasets before release. The same Datadog tracer can be used from development through production, with tracing across requests, services, agent reasoning, upstream and downstream dependencies, and end-user experience. It also provides alerting, role-based access control, sensitive-data protection, and HIPAA compliance controls. The product addresses agent quality, cost, safety, and operational behavior in addition to conventional application signals such as latency, errors, traffic, and saturation.
Google DeepMind is an artificial-intelligence research and development organization that builds AI models and systems for general use, science, robotics, speech, image and video generation, music, and coding. Its featured systems include the Gemini family, Gemma open models, Veo video generation, Lyria music generation, Gemini Robotics, and AI systems for scientific discovery. The site also publishes research and provides access to models through products and developer platforms such as Google AI Studio, the Gemini API, and related enterprise tools.
An application programming interface from xAI that provides programmatic access to Grok language models, including Grok 2, for generating and analyzing text through software applications.
Intercom is a customer messaging and engagement platform that provides businesses with live chat, in-app messaging, help desk tools, automation, and conversational bots. It is developed and offered by Intercom, a software company for customer support and product growth.
Zendesk is a customer service software company that provides a cloud-based service platform for support, ticketing, and customer engagement. The platform includes tools for help desks, knowledge bases, omnichannel messaging, and AI-powered automation. It is operated by Zendesk, Inc.
Freshdesk is a cloud-based customer support and helpdesk platform developed by Freshworks. It provides ticketing, multichannel support (email, phone, chat, social), automation, self-service portals, SLA management, reporting and integrations for businesses.
MLX is an array framework for machine learning on Apple silicon, developed by Apple machine learning research. It provides Python, C++, C, and Swift APIs modeled on NumPy, plus higher-level packages such as mlx.nn and mlx.optimizers with PyTorch-like interfaces. Its composable transformations support automatic differentiation, automatic vectorization, and computation-graph optimization; computations are lazy, graphs are built dynamically, and operations can run on supported CPUs and GPUs using shared unified memory rather than explicit data transfers. MLX is distributed through PyPI and supports model training and deployment workflows including language-model training and generation, LoRA fine-tuning, image generation, and speech recognition.
Plaud is an AI-powered note-taking platform that consolidates in-person conversations, phone calls, and online meetings into a single workflow, offering automated transcription and summaries. The service is presented at plaud.ai and the site states it has over 2.5 million users.
Workers for Platforms is a Cloudflare offering that lets platform operators embed the Cloudflare Workers serverless runtime into their products, enabling customers to run user-provided server-side code at Cloudflare's edge. It is provided and maintained by Cloudflare and builds on the Cloudflare Workers ecosystem for distributed compute.
OpenTimestamps is a timestamping proof standard and software ecosystem for proving that data existed before a particular point in time. It defines operations for creating and independently verifying timestamp proofs; the supported blockchain timestamping method is Bitcoin, while the format is designed to accommodate other methods. Users can hash and stamp files in a browser or use client tools that submit requests to calendar servers, which later provide blockchain attestations. The ecosystem includes Python, JavaScript, Java, and Rust libraries or clients, a calendar server, browser-based stamping and verification, and free default calendar services that do not require registration or an API key.
Common Expression Language (CEL) is an embeddable expression language and specification for interoperable expression evaluation across applications, programming languages, and platforms. Its syntax resembles expressions in C, C++, Java, and JavaScript, and it is designed for uses such as security policies and protocol data. A CEL system compiles a textual expression into an abstract syntax tree and binary representation, evaluates that representation against a context containing typed variables such as protocol-buffer messages, and usually produces a Boolean result. The embedding application supplies functions and data through the evaluation context. CEL is mutation-free, not Turing-complete, and designed for linear-time evaluation; type-checked expressions can be serialized as canonical protocol buffers for persistence or cross-process communication. The specification and accompanying tooling are released under the Apache License.
OpenGrok is a Java-based source code search, browsing, and cross-reference engine developed by Oracle. It indexes source trees so users can search and navigate code, understand various programming-language file formats, and cross-reference version-control histories from multiple source-code management systems. It can be deployed as a web application or run in a Docker container. The project is distributed under the license stated in its repository and uses semantic versioning; major and minor updates may require reconfiguration or a full reindex, while micro updates require redeploying the web application.
Trace is an AI-powered browser developed by Breadcromb that creates a private knowledge layer by unifying tabs, documents, and conversations and uses that context to assist the user. The product is described as local-first by default and is offered as a free download.
Open-source Python toolkit that rewrites AI-generated text to read human-written. Ships four documented methodologies — translation chaining, multi-turn LLM rewriting, detection-guided feedback loops, and mixed-engine translation — plus a Standard Pipeline that chains two LLM rewrite passes (DeepSeek/OpenRouter, second pass carries the first as history) with two cross-engine translation hops (EN → ZH → JA → FI → EN across different engines), so no single engine's structural fingerprint survives. Claims 100% key-information retention on 50 verified text pairs; the repo warns detector scores are probabilistic and against misrepresenting authorship.
Awesome MCP Servers is a curated collection of Model Context Protocol (MCP) server implementations, maintained as a GitHub repository with a synchronized web directory. It catalogs production-ready and experimental servers that let AI models interact with local and remote resources through standardized interfaces, including file access, databases, API integrations, browser automation, cloud platforms, research services, and other contextual tools. The repository organizes entries by capability, programming language, deployment scope, operating system, and whether an implementation is official.
Checkstyle is an open-source Java code-quality and static-analysis tool that checks source code against coding standards and configurable best-practice rules. It supports the Google Java Style Guide and Sun Code Conventions by default, and its checks can be configured in XML, including rules such as detecting fall-through between branches of a switch statement. It can be run as a command-line program or an Apache Ant task, and can be added to builds through Maven Central. The project is developed and maintained by volunteers and is licensed under the GNU LGPL v2.1.
An open-source set of reinforcement-learning training environments for Pollen Robotics' Microduck bipedal robot, built on mjlab with MuJoCo Warp and PPO. It trains walking, standing, recovery, sitting, ground-picking, kicking, rolling, and roller-skating policies, with flat, rough, slope, and backlash variants where applicable. The environments encode a sim-to-real pipeline using BAM actuator physics for Dynamixel XL330 servos, domain randomization for battery voltage, voltage sag, command delay, and friction, and simulated gear backlash with output-side encoder feedback. Policies use a shared 61-dimensional observation contract for runtime hot-swapping, are trained at 50 Hz, exported to ONNX with the observation normalizer embedded, and deployed by the Microduck runtime. Training runs locally through a CUDA GPU or can be submitted to Hugging Face Jobs; the project is licensed under Apache 2.0, while its 3D model files use CC BY-SA-NC.
Corsair is an open-source product integration platform for connecting users and services to third-party applications. It exposes a REST API and a common syntax across integrations, allowing the same integration layer to support AI agents, backend services, and multi-tenant dashboards instead of requiring separate integration-specific glue code. Corsair maintains adapters for the integrations, while users can self-host the platform or use Corsair Hub for managed OAuth token refresh and webhooks. The project is licensed under Apache License 2.0.
WandEnhancer is an open-source interoperability and user-interface enhancement tool for the Wand (WeMod) desktop application. Its .NET patcher modifies files in a selected local Wand installation, while a bundled version.dll proxy changes Wand's Electron ASAR-integrity fuse byte inside the application's own process. It supports local configuration management, client-version compatibility adjustments, client-side layout and theme customization, AI features, and custom JavaScript injection into Wand's renderer. Injected scripts have DOM and Node require access, run with the Wand client's privileges, and are wrapped so thrown errors are logged rather than crashing the application. The tool includes an optional Remote Web Panel that starts a plain HTTP/WebSocket server on the local network at TCP port 3223, allowing a phone on the same LAN or VPN to control the active trainer. The panel has no pairing code, so the project warns against exposing the port to the public internet. The patching step has no updater or project telemetry, although the optional panel can use Wand's existing API/CDN paths. The project does not publish prebuilt executables; users build it on Windows with CMake, Node.js and pnpm, Visual Studio/MSBuild, and .NET Framework 4.8, or run its GitHub Actions workflow from a fork. It is licensed under Apache-2.0.
ODS is a self-hosted local AI server stack developed by Osmantic for Linux, Windows with WSL2 and Docker Desktop, and Apple Silicon macOS. Its installers detect the host hardware, select an installable model, generate credentials, and launch a preconfigured stack containing local LLM inference, Open WebUI, a management dashboard, LiteLLM, embeddings, speech-to-text, text-to-speech, agents, n8n workflows, Qdrant-based RAG, search, ComfyUI image generation, and operational and privacy tools. Local mode is the default; cloud and hybrid modes can route through OpenAI, Anthropic, or Together APIs. The stack uses a bootstrap model so chat can begin while the hardware-selected full model downloads in the background. A manifest-based extension system discovers service folders and merges their Docker Compose definitions into the stack, while the `ods` CLI manages services, models, modes, configuration, presets, and lifecycle operations. The repository documents GPU and unified-memory paths for NVIDIA, AMD, Intel Arc, and Apple Silicon hardware, along with model selection, recovery, validation, and operator workflows. ODS is distributed as an Apache-2.0-licensed open-source repository and can be installed with shell or PowerShell installers. The repository identifies version 2.6.0 as its stable release and states that no cloud account or subscription is required for local operation.
User Scanner is an open-source command-line and Python library OSINT suite for investigating email addresses and usernames across hundreds of platform modules. It checks account registrations and scrapes profile metadata such as avatars, biographies, follower counts, user IDs, seller statuses, and other account attributes. The tool supports cross-scan pivoting: initial results can yield handles, profile links, aliases, or exposed email addresses that are used as inputs for subsequent scans. It also provides username permutation generation, Hudson Rock infostealer-breach lookups, parallel HTTP scanning with TLS fingerprint impersonation, proxy rotation and validation, category or module targeting, bulk-file input, and PDF, JSON, and CSV exports. It can be installed from PyPI or run through Nix. An optional Model Context Protocol server exposes username scanning, email scanning, and module-discovery tools to compatible AI clients. The project describes itself as intended for educational use, authorized security research, and defensive OSINT investigations.
Invidious is an open-source alternative front end to YouTube. It provides video playback, subscriptions independent from Google, channel notifications, audio-only playback, customizable themes and homepages, multilingual support, data import and export, embedded videos, and a developer API. It does not use the official YouTube APIs and can be used without JavaScript; users can access it through public instances or host it themselves. The project is distributed under the GNU Affero General Public License v3.
video-use is an open-source video-editing tool for coding agents such as Claude Code, Codex, Hermes, and Openclaw. It takes raw footage in a folder and produces an edited MP4, with support for removing filler words and dead space, applying FFmpeg color-grading chains, adding short audio fades at cuts, burning customizable subtitles, and generating animation overlays through HyperFrames, Remotion, Manim, or PIL. Its pipeline transcribes each source with ElevenLabs Scribe, packing word-level timestamps, speaker diarization, and audio events into a Markdown view. The agent reasons over that transcript and requests filmstrip, waveform, and word-label composites from timeline_view only at decision points, rather than processing every video frame as visual context. It then creates an edit decision list, renders the result, evaluates the rendered video at every cut boundary for visual jumps, audio pops, and subtitle problems, and can revise and re-render up to three times. The tool uses an approval-oriented workflow in which the agent inventories sources, proposes an editing strategy, waits for confirmation, executes the edit, evaluates it, and persists session memory in project.md. It requires FFmpeg and an ElevenLabs API key; yt-dlp is optional for downloading online sources. Outputs are placed in an edit directory next to the source footage.
Awesome DESIGN.md is a curated collection of DESIGN.md files analyzing the design systems of developer-focused websites and other brands. Each plain-text Markdown file records visual themes, semantic color tokens, typography, component styles and states, layout, elevation, responsive behavior, design guardrails, and prompts for AI design or coding agents; accompanying preview files show the documented tokens and components on light and dark surfaces. The collection follows the DESIGN.md format introduced by Google Stitch. A file can be copied into a project so an AI agent can use the documented design language when generating UI, without Figma exports, JSON schemas, or special parsing tools. The repository is maintained through contributions, issue reports, and pull requests; contributors are asked to open an issue before proposing changes, and the files are provided under the MIT License as-is.
ReClip is a self-hosted, open-source video and audio downloader with a web interface. It accepts one or more URLs from sites supported by yt-dlp, fetches video information and thumbnails, and downloads the selected content as MP4 video or MP3 audio using yt-dlp and ffmpeg. The interface supports quality or resolution selection, bulk downloads, and automatic URL deduplication. The backend uses Python and Flask, while the frontend is a single file of vanilla HTML, CSS, and JavaScript with no build step. It can run locally from a shell script or in Docker, and is distributed under the MIT license.
Sequoia-X is a Python quantitative stock-selection system for China A-share markets. It uses an object-oriented architecture, vectorized calculations, and incremental data updates to scan for technical patterns after the market close and send results to a Feishu group. The data engine retrieves historical and incremental adjusted daily K-line data from Baostock, stores it in a local SQLite database, and supports a backfill mode for initially loading market history and a daily mode for incremental updates, strategy execution, and notification. Its included strategies cover Turtle breakout, moving-average and volume breakout, high-tight-flag breakout, limit-up shakeout, uptrend limit-down reversal, and RPS relative-strength breakout. The project requires Python 3.10 or later and is distributed under the MIT license.
TypeWords is an open-source English vocabulary and article practice tool developed as a Nuxt application. It supports follow-along, dictation, self-test, and spelling-from-memory modes for words, with a smart mode that selects learning words using memory-curve calculations. Word entries can include phonetics, American and British pronunciation, example sentences, phrases, synonyms, roots, etymology, and error statistics; incorrect words, favorites, and mastered words can be stored for review or automatically skipped. Users can also practice built-in or imported articles through sentence-by-sentence follow-along and dictation modes, including bilingual comparison and automatic pronunciation. The project can run locally with data saved in the browser or local environment, and is distributed under an open-source repository.
Portless is a command-line development proxy that replaces numbered localhost ports with stable, named local URLs such as https://myapp.localhost. It runs an application command, assigns the child process an available port through the PORT environment variable, registers the app with an HTTP/2 HTTPS proxy, and routes requests from the named hostname to that process. It can infer and run package.json development scripts, inject port and host flags for frameworks that do not honor PORT, manage multiple workspace packages, and register aliases or subdomains. The proxy supports custom TLDs, local certificate generation and trust, WebSockets, LAN access through mDNS, startup services, and optional sharing through Tailscale or ngrok. Portless is distributed as an npm package for macOS, Linux, and Windows and is pre-1.0.
fanqiang is a curated collection of internet-circumvention tools, tutorials, and related resources maintained in a GitHub repository. It curates guides and software for Windows, macOS, Linux, Android, iOS, routers, browsers, game consoles, and Apple TV, including instructions for configuring V2Ray, Shadowsocks, SSR, Clash, Tor Browser, and other proxy tools. The repository also includes bundled browser packages and guides for setting up self-hosted V2Ray or Shadowsocks servers, and states that its tool collection is periodically updated.
prompts.chat is a curated collection and community-maintained library of prompts for AI chat models, originally known as Awesome ChatGPT Prompts. Users can browse prompts, submit additions through prompts.chat, and access the collection as a website, CSV file, Markdown document, or Hugging Face dataset; submitted prompts sync to the repository automatically. It supports use with ChatGPT, Claude, Gemini, Llama, Mistral, and other modern AI assistants. The project also provides an interactive prompt-engineering book, a game-based prompting experience for children, a CLI, a Claude Code plugin, and an MCP server. Its open-source application can be self-hosted with custom branding, themes, authentication, and PostgreSQL; the repository documents setup through `npx prompts.chat new` or a manual clone and installation. Source code and site-authored content use the MIT License, while prompt content and data are dedicated to the public domain under CC0 1.0 Universal.
Exploitarium is a curated collection of public proof-of-concept exploits and vulnerability-research writeups maintained as a GitHub repository. It organizes former standalone PoC repositories into self-contained folders that preserve their original READMEs and tracked files, while new research entries are added directly to the archive. The repository documents a consolidation process that verifies paths, Git object types, tree modes, and blob IDs for migrated files, and states that the material is intended for good-faith, open-disclosure research rather than malicious use.
NoSignups is a curated directory of open-source browser tools that can be used without accounts, email addresses, or tracking. It organizes entries into categories such as productivity, design, development, writing, privacy, utilities, data, media, and education, with records containing a tool name, description, direct URL, category, tags, source repository, and optional license and recommendation metadata. The directory is implemented as a React and TypeScript project, accepts community submissions, and is released under the GPL-3.0 license; listed tools retain their own licenses.
Everything Claude Code is a Claude Code plugin and toolkit containing configurable agents, skills, slash commands, rules, hooks, contexts, MCP server configurations, scripts, and tests for AI-assisted software development. Its agents delegate tasks such as planning, code review, security review, testing, and build-error resolution. Skills define workflows including test-driven development, security review, continuous learning, and verification; hooks run on Claude Code tool or session events to load and save context, suggest compaction, and evaluate sessions; and rules provide persistent guidance for security, coding style, testing, Git workflow, delegation, and performance. Node.js scripts support Windows, macOS, and Linux and detect a project's preferred package manager from environment settings, project configuration, package.json, lock files, or global configuration. The repository can be installed as a Claude Code plugin or copied into Claude configuration directories manually. It includes a Node.js test suite and is distributed under the MIT license.
nvm is a POSIX-compliant shell-based version manager for Node.js, designed for per-user installation and per-shell use. It installs, downloads, compiles, switches between, and removes Node.js versions, with support for aliases, LTS release lines, io.js, system Node.js, binary mirrors, offline installation, and migration of global npm packages. A project's .nvmrc file can specify a version that nvm uses when running commands in that project or its subdirectories. nvm is distributed as a sourced Bash script rather than an executable binary and supports POSIX shells including sh, dash, ksh, zsh, and bash on Unix, macOS, and Windows WSL; the repository also documents Docker and shell-completion setups.
Marketing Skills is a collection of Markdown-based skills for AI coding agents, built by Corey Haines. It provides specialized workflows for conversion optimization, copywriting, SEO, analytics, advertising, growth engineering, sales, and related marketing tasks, and supports Claude Code, OpenAI Codex, Cursor, Windsurf, and agents implementing the Agent Skills specification. The skills share context and cross-reference one another. The product-marketing skill serves as the foundation, with other skills consulting it for product, audience, and positioning information before handling tasks such as CRO, copywriting, SEO audits, analytics setup, email flows, pricing, or sales enablement. Skills can be installed individually or as a collection through the Skills CLI, Claude Code's plugin system, a repository clone or submodule, or SkillKit. The repository is built and maintained through community contributions, with issues and pull requests welcomed. It is free and MIT-licensed.
Stremio Web is the official web UI for Stremio, a media center for discovering and playing movies, series, and channels through addon-provided catalogs. It provides library and Continue Watching synchronization through a Stremio account, Chromecast casting, addon-provided or local subtitles, keyboard-controlled playback, community translations, and installation as a standalone progressive web app. The interface is a React application that exchanges state with stremio-core, a Rust engine compiled to WebAssembly and run in a Web Worker. The core communicates with the Stremio API and addons, while playback is handled through stremio-video, which selects the player implementation for the environment. The project is developed by Smart Code OOD, can be run locally with Node.js and pnpm or in Docker, and is released under the GPL-2.0 license.
Camofox Browser is an open-source anti-detection headless browser server for AI agents, built by the team behind jo and powered by Camoufox, a Firefox fork that applies fingerprint spoofing at the C++ level. It exposes browser automation through a REST API with accessibility snapshots, stable element references for clicking and typing, navigation and search macros, screenshots, downloads, file uploads, structured extraction, YouTube transcript extraction, and OpenAPI documentation. Browser contexts provide isolated cookies and storage, with optional persistence, cookie import, proxy and GeoIP configuration, VNC-based interactive login, and Playwright session tracing. It can run from npm, Docker, Fly.io, or Railway, and supports lazy browser startup and idle shutdown. The repository is licensed under MIT.
LunaTV is a self-hosted, cross-platform streaming aggregator and video player. The repository README also labels the application MoonTV. It provides multi-source search, detail pages with episode and metadata information, browser playback through ArtPlayer and HLS.js, favorites and watch-progress synchronization, responsive web layouts, and a progressive web app mode. It can optionally skip segmented video advertisements as an experimental feature. LunaTV is built with Next.js, Tailwind CSS, and TypeScript and is distributed for Docker-based deployment. A fresh deployment contains no built-in playback or live-stream sources; an administrator must supply Apple CMS V10-compatible source APIs through the configuration interface. Redis, Kvrocks, and Upstash can store favorites and viewing history, and the project documents compatibility with the Selene mobile client and OrionTV on Android TV. The repository page states a CC BY-NC-SA license with restrictions on commercial use, while the README's license section states MIT, so the licensing information is inconsistent in the supplied page text.
DeerFlow is an open-source long-horizon super-agent harness developed in the ByteDance repository. Built on LangGraph and LangChain, it orchestrates a lead agent with sub-agents, progressively loaded skills, memory, tools, and sandboxed execution environments for research, coding, file work, report generation, slide creation, web-page creation, and other multi-step tasks. The runtime provides a filesystem for uploads, workspaces, and outputs; configurable web search, web fetching, browser control, Bash and file tools; MCP servers; reusable Markdown-based skills; long-term memory; scheduled tasks; and messaging-app channels. Sub-agents receive scoped context, tools, and termination conditions, and can run bounded parallel work before returning structured results to the lead agent. Sandbox modes include local execution, isolated Docker containers, Kubernetes-backed execution, and E2B-based environments, depending on configuration. DeerFlow can run locally or with Docker Compose and exposes a Web UI and Gateway at the documented local address. Its setup wizard generates configuration for model providers, search, sandbox, Bash, and file-writing options; the project also documents integrations for LangSmith, Langfuse, and Monocle tracing, personal access tokens, custom agents, MCP background tasks, and extensible Python packages. Version 2.0 is a ground-up rewrite of the earlier 1.x Deep Research framework; active development is on the 2.0 line.
Lightpanda Browser is a headless browser built from scratch in Zig for AI agents and web automation. It is not based on Chromium, Blink, or WebKit; its implementation includes an HTML parser, V8 JavaScript support, a DOM tree and APIs, network loading, cookies, form input, proxy support, network interception, and optional robots.txt compliance. It can dump pages as HTML, Markdown, PNG, or PDF, or expose a Chrome DevTools Protocol and WebDriver BiDi server for clients such as Puppeteer and Playwright. Its native agent mode uses an LLM to navigate pages, click through flows, fill forms, and extract structured data. Sessions can be exported as deterministic, token-free PandaScript JavaScript and replayed without a model at runtime. The browser also provides an MCP server over stdio or HTTP, with isolated or shared browsing sessions for multiple agents. The project distributes nightly binaries for Linux and macOS, Docker images for Linux amd64 and arm64, and packages for Homebrew and the Arch User Repository. It can also be built from source with Zig and its V8, Libcurl, and html5ever dependencies.
Andrej Karpathy Skills is a set of guidelines for Claude Code and Cursor that configures coding agents to avoid common LLM coding pitfalls. Its four principles are Think Before Coding, which requires explicit assumptions, clarification of ambiguity, and stated trade-offs; Simplicity First, which limits speculative features and unnecessary abstractions; Surgical Changes, which restricts edits to the requested scope; and Goal-Driven Execution, which turns tasks into verifiable success criteria and test-driven loops. The guidelines can be installed as a Claude Code plugin, copied into a project's CLAUDE.md file, or applied through the repository's Cursor rule. The repository is licensed under the MIT License.
Browser Use is an AI browser-automation framework and hosted service for assigning web tasks to agents. Its agents operate websites through a browser by opening pages, clicking controls, entering text, filling forms, and extracting structured data. The project provides a Python library for embedding browser agents in applications, a CLI for agent-driven one-off tasks, and a Browser Use Cloud API with hosted agents and managed browser infrastructure. The library supports different LLM providers, custom tools, structured output, reusable browser profiles, and parallel or scheduled automation. The open-source library is free to use and licensed under the MIT License; the hosted service adds scalable browser infrastructure, proxy rotation, stealth browser fingerprinting, CAPTCHA handling, persistent storage, and integrations.
Escrcpy is an open-source graphical interface for displaying and controlling Android devices through the scrcpy engine. It provides an embedded mirror with screen-fitting and orientation handling, configurable mappings for touch, joystick, swipe, scrolling, and automation, simultaneous control of multiple devices with broadcast input, batch screenshots and APK installation, and a draggable control bar for device operations. It also supports wireless ADB connections with LAN discovery and Gnirehtet reverse tethering, visual step-by-step automation with screen-image recognition, global shortcuts, and an MCP-based AI assistant with multi-model chat. Released packages are available for manual installation, with macOS installation through Homebrew; advanced features from the private EscrcpyX extension are paid, and the project states that updates are irregular.
Vulture is a Python static-analysis tool that finds unused and unreachable code. It parses Python files with the ast module, records defined and used object names while traversing their abstract syntax trees, and reports definitions that are not used; it also detects code after return, break, continue, or raise statements and unsatisfiable conditional or loop conditions. It analyzes Python files and directories, assigns reported items confidence values, supports minimum-confidence filtering, sorts findings by size, and can generate Python whitelist files to suppress false positives caused by dynamic usage. It can be installed with pip, configured through pyproject.toml, run as a pre-commit hook or GitHub Action, used through a VS Code extension, or called programmatically. The repository is maintained by Jendrik Seipp and is distributed as a Python project.
TradingAgents is an open-source, multi-agent LLM financial trading framework developed by TauricResearch for research. It models a trading firm through specialized agents: fundamental, sentiment, news, and technical analysts produce market assessments; bullish and bearish researchers debate those assessments; a Trader Agent proposes the timing and size of a trade; and risk-management and portfolio-management agents evaluate and approve or reject it before execution on a simulated exchange. The framework uses LangGraph and can be run through an interactive CLI, Docker, or as a Python package via the TradingAgentsGraph class and its propagate function. It accepts market and ticker inputs, analysis dates, configurable research depth and debate rounds, and LLM backends including hosted providers, Ollama, and other OpenAI-compatible servers. It gathers financial, market, news, macroeconomic, and social-sentiment data, and supports persistent decision logs and optional LangGraph checkpoint recovery. TradingAgents is intended as a research scaffold rather than financial, investment, or trading advice. Results can vary with the language model, sampling settings, data sources, analysis period, and other nondeterministic factors.
LLM Wiki is a cross-platform desktop application maintained by nash_su that turns imported documents into a persistent, organized, interlinked knowledge base. It uses a three-layer structure of immutable raw sources, LLM-generated wiki pages, and configuration rules, with ingest, query, and lint operations; generated pages use YAML frontmatter, source traceability, an index, an operation log, and [[wikilink]] cross-references. Its ingest pipeline first analyzes each source for entities, concepts, connections, contradictions, and recommended structure, then generates or updates wiki pages, summaries, indexes, research queries, and human-review items. SHA256 caching skips unchanged files, while a persistent queue supports serialized processing, retries, cancellation, and crash recovery. It accepts PDFs, Office documents, EPUB/MOBI, Org mode, images, media, web clips, and URL batches, with optional vision-based image captions and MinerU processing for complex PDFs. Retrieval combines tokenized keyword search, optional LanceDB vector search through OpenAI-compatible embedding endpoints, and graph expansion based on direct links, shared sources, Adamic-Adar similarity, and page-type affinity. The application also provides Louvain community detection, graph insights, deep research through Tavily, SerpApi, or SearXNG, a Rust tool-using chat agent, review workflows, Mermaid and KaTeX rendering, a Chrome web clipper, and project export/import. It includes a token-protected local HTTP API and MCP server for searching, reading files, traversing the graph, rescanning sources, and calling the backend agent. The application is built with Tauri v2, React, TypeScript, and Rust, has macOS, Windows, and Linux builds, and is licensed under GPL-3.0.
Skills is a command-line tool for the open agent skills ecosystem. It installs reusable instruction sets defined in SKILL.md files into supported coding agents such as OpenCode, Claude Code, Codex, Cursor, and others. The `npx skills` CLI resolves skills from GitHub, GitLab, other Git URLs, local paths, or direct file and archive URLs; lists available skills, selects individual skills, and installs them project-wide or globally by symlink or copy. It can also generate a prompt for a skill without installing it, launch a supported agent with that prompt, search for skills, update or remove installations, and create new SKILL.md templates. Skills use YAML frontmatter with a name and description, and the CLI discovers them in standard agent and repository skill directories. The project is licensed under the MIT License.
QuantsPlaybook is a GitHub resource containing Python reproductions of quantitative-investment research reports from Chinese securities firms. It covers market-timing strategies, factor construction, quantitative value methods, and portfolio optimization, with more than 100 listed strategy implementations including RSRS and QRS timing, volatility and technical-analysis methods, multi-factor models, FFScore, cash-flow selection, differential-evolution portfolio optimization, and deep multi-task time-series momentum. The repository organizes each strategy with related research papers or brokerage reports and Python files containing notebook-based reproductions; its workflows cover data acquisition, analysis, backtesting, visualization, and, where used, machine-learning or signal-decomposition methods. It depends on data sources such as JQData and Tushare and references tools including pandas, NumPy, Qlib, Backtrader, PyTorch, TensorFlow, LightGBM, XGBoost, scikit-learn, and Jupyter.
Shadowrocket-ADBlock-Rules-Forever is a maintained collection of routing and ad-blocking rule files for the iOS Shadowrocket app. It provides configurations that determine which domains use a proxy or direct connection, including blacklist, whitelist, domestic/international split, return-to-China, global proxy, global direct, and ad-block-only variants. The project generates its rules automatically with Python from sources including GFWList, Greatfire Analyzer blocked-domain data, EasyList, EasyList China, Peter Lowe's ad and privacy-tracking lists, 乘风 rules, and project-specific rules. It also provides separate files for iOS web, app, and video advertising, plus optional lazy configurations with proxy groups. The repository states that published rules are rebuilt daily at 08:00 Beijing time and can be downloaded or automatically updated through Shadowrocket and iOS Shortcuts. The rules are plain-text configuration files and do not themselves provide proxy access.
QLScriptPublic is a public repository of JavaScript automation scripts for the Qinglong panel, maintained by the GitHub user smallfawn. The scripts automate sign-ins, daily tasks, points collection, and related account actions for numerous services, brands, apps, and WeChat mini programs. WeChat mini-program scripts require configuring `wx_server_url` and `wx_auth` to obtain mini-program codes, and generally require prior registration, authorization, or phone-number binding in the relevant mini program. The repository can be loaded into Qinglong with `ql repo https://github.com/smallfawn/QLScriptPublic.git`. Its disclaimer describes the scripts as learning, testing, and research material, prohibits commercial use and redistribution, disclaims responsibility for legality, accuracy, script errors, privacy issues, and resulting losses, and asks users to delete downloaded content within 24 hours.