2,424 tools and products — trending open source, and what gets used in AI and other work.
Franzy is an online platform that helps prospective franchisees discover, compare, and connect with franchising opportunities. The service lists over 1,000 franchise options and provides resources to begin the franchise search and contact franchisors.
Codecrafters is an online platform offering advanced programming challenges for experienced engineers. It provides guided projects to rebuild real production tools (for example Redis, Git, and SQLite) from scratch inside the user's IDE.
A visa and immigration service that helps international founders assess their eligibility for United States visas and relocate to the country.
The C100 is a nonprofit membership organization of Canadian founders, executives, and investors based in the San Francisco Bay Area and other global tech hubs that connects Canadian technology companies to mentors, investors, design partners, and senior executives. It provides introductions, programming, mentorship and networking events to help Canadian startups access U.S. markets and scale internationally. The organization operates as a member-run network rather than a commercial service.
RAGFlow is an open-source retrieval-augmented generation (RAG) engine from infiniflow that provides a context layer between documents and large language models, combining document understanding with agent capabilities. It ingests heterogeneous sources including Word, Slides, Excel, text files, images, scanned documents, structured data, and web pages; extracts knowledge from complex formats; and uses configurable template-based chunking, multiple retrieval methods, and fused re-ranking to assemble context for language-model applications. Its workflows provide traceable citations and visualized text chunks for human review, and it supports configurable language and embedding models, agent templates, orchestration, APIs, and self-hosting. A hosted cloud service is also available.
Unsloth is an open-source local desktop and web tool for running, training, fine-tuning, and deploying language, diffusion, embedding, audio, and vision models. It supports Windows, macOS, Linux, WSL, multi-GPU setups, NVIDIA, AMD, and Intel GPUs, CPUs, and the Vulkan backend. The project is distributed as Unsloth Desktop, Unsloth Studio, and Unsloth Core. It can run local models, provide an OpenAI-compatible API, connect models to agents and MCP tools, perform search and RAG, and offer remote or LAN access. Its training functions cover reinforcement learning, LoRA, QLoRA, full fine-tuning, pretraining, GRPO, DPO, and FP8, with export formats including GGUF, NVFP4, and FP8. The repository claims that its fine-tuning workflows use two times less training time and 70% less VRAM.
Paritok is an open-source compression gateway for AI coding agents. It runs as a drop-in proxy between an agent and an upstream LLM API, stripping irrelevant tool schemas, compressing tool results and file reads, and summarizing stale conversation history before forwarding requests. Compressed content remains recoverable on demand through the gateway, while responses pass back to the agent unchanged. It supports agents such as Claude Code, Cursor, Codex, and OpenHands, as well as agents that support BASE_URL configuration, and is powered by an open-source code-focused 4B compression model.
Freebuff is an ad-supported AI coding-agent suite from CodebuffAI. Its CLI runs in a project from the terminal, where specialized agents find relevant files, plan and implement changes, run project checks, and review the results. The suite also includes desktop, web, cloud, and chat products for running parallel local agents, building full-stack applications, operating on GitHub repositories, and conducting research. Freebuff provides a catalog of included AI models without a subscription, credits, or an API key; text ads support access to those models. Desktop can run concurrent agents in separate workspaces, while the web and cloud products provide hosted sandboxes, previews, terminals, browser-based research, and deployment workflows.
An open-source authentication library for JavaScript and TypeScript web applications maintained by the Auth.js project. It implements OAuth/OpenID Connect provider sign-in, email and credential-based authentication, session management (JWT or database-backed sessions), and provides adapters to persist accounts and sessions. Auth.js runs on Node.js and is commonly used with Next.js and other Node web frameworks; source code is published on GitHub and the project is distributed under the MIT license.
Snorkel is a system for programmatically building and managing machine-learning training data through weak supervision. It encodes expert knowledge as labeling functions, uses those functions to generate labels at scale, and supports data labeling, augmentation, slicing, and management for adaptable training datasets. The project began at Stanford and is now complemented by Snorkel Flow, an end-to-end AI application development platform based on the same ideas.
Mercury 2 is a diffusion language model developed by Inception that generates multiple tokens in parallel through iterative refinement rather than decoding strictly one token at a time. It is designed for latency-sensitive applications and has been reported to achieve speeds exceeding 1,000 tokens per second.
Harbor is a framework from the creators of Terminal-Bench for evaluating and optimizing AI agents and language models in container environments. It can run arbitrary agents, including Claude Code, OpenHands, and Codex CLI, against shared or custom benchmarks and environments such as Terminal-Bench, SWE-Bench, and Aider Polyglot. Harbor launches benchmark runs locally with Docker or in parallel through cloud and sandbox providers, and can generate rollouts for reinforcement-learning optimization. It is distributed as a Python package installable with uv or pip and provides command-line tools for running datasets, selecting agents and models, and listing supported benchmarks.
An AI tool that orchestrates agents to synthesize realistic reinforcement-learning environments and tasks from domain descriptions, real usage data, scenarios, and personas.
Workday, Inc. is a US-based enterprise software company that provides cloud-native human capital management (HCM), financial management, payroll, talent management, planning, and analytics applications delivered as multi-tenant SaaS. Workday tenant instances ingest employee records, payroll and financial transactions, time and absence data, recruiting and compensation inputs and produce HR and finance ledgers, payroll runs, regulatory filings, operational reports and dashboards; integrations and data exchange use Workday Studio, Enterprise Interface Builder (EIB), SOAP/REST APIs and prebuilt connectors for SSO and ERPs. Founded in 2005 by Aneel Bhusri and Dave Duffield, Workday is commercially licensed enterprise software that includes machine-learning features and an assistant/chat capability but is primarily an enterprise SaaS platform.
Tether is a company (Tether Limited) that issues USDT, a fiat-collateralized stablecoin pegged to the US dollar. USDT tokens are issued on multiple blockchains (originally the Bitcoin Omni layer and later on Ethereum as ERC-20, Tron as TRC-20, Solana and others), are used as a dollar-referenced medium for trading and settlement across exchanges and wallets, and Tether's reserves and audit practices have been the subject of regulatory scrutiny and periodic attestations by accounting firms.
Ledger is a hardware cryptocurrency wallet and digital-asset management platform. Its hardware signers are used to manage and authorize transactions for cryptocurrencies including Bitcoin, Ethereum, Solana, XRP, stablecoins and other supported assets, while Ledger Wallet provides functions such as buying, swapping, staking, spending and portfolio tracking. The platform also compares rates from cryptocurrency providers across multiple chains and offers services including Ledger Recover.
HaGeZi DNS Blocklists is a maintained collection of DNS filtering lists for network-wide blocking of ads, affiliate links, trackers, telemetry, phishing, malware, scams, cryptojacking, and other unwanted or malicious domains. It provides Multi lists in Light, Normal, Pro, Pro++, and Ultimate variants, along with specialized lists for fake sites, pop-up ads, threat intelligence, newly registered domains, dynamic DNS abuse, URL shorteners, gambling, social networks, adult content, native device trackers, and other categories. Lists are distributed in formats compatible with Pi-hole, AdGuard, DNSMasq, RPZ, common DNS servers, ad blockers, and content blockers.
Helical Insight is an open-source business intelligence and embedded analytics platform for self-hosted, cloud, on-premises, and hybrid deployments. It combines natural-language, chat-driven AI analytics with bring-your-own-LLM support, automatic SQL generation, report summarization, and insight discovery; interactive dashboards with filters, KPI widgets, drill-down and drill-through interactions, maps, charts, pivot tables, and custom JavaScript visualizations; and paginated, pixel-perfect reports for invoices, financial statements, operational reports, and regulatory reporting. Dashboards, reports, and the AI chatbot can be embedded in web applications, SaaS platforms, customer portals, and enterprise applications. Reports and dashboards can be exported to PDF, Excel, CSV, Word, HTML, JSON, and XML, scheduled for delivery by email, and distributed as personalized reports. The platform supports white labeling, multi-tenancy, row-, column-, and table-level security based on user context, and SSO through JWT, Okta, Keycloak, OAuth, and custom token-based mechanisms. It connects to relational databases, NoSQL and big-data systems, flat files, cloud storage, and APIs through native connectors, JDBC, REST APIs, custom integrations, and uploaded JDBC drivers. Docker, Kubernetes, caching, pagination, virtualization, load balancing, clustering, and REST APIs support deployment and extension, while developers can customize it with Java, JavaScript, CSS, HTML, Liquid Template Language, plugins, and custom workflows. The project states that its enterprise features are available in the free open-source version.
BackUp Maker is a Windows backup application from ASCOMP Software (Germany) that creates scheduled backups of files and folders to local drives, network shares, FTP servers and removable media. It supports mirror-style and incremental/point-in-time backups with versioning, optional encryption of backup archives, and scheduling/automation features.
R is an open-source programming language and software environment for statistical computing and graphics, originally created by Ross Ihaka and Robert Gentleman and now maintained by the R Core Team and the R Foundation. It reads and writes common data formats (CSV, Excel via packages, SPSS/SAS/Stata via packages), provides facilities for data manipulation, statistical modeling, hypothesis testing, time series and high-quality plotting (base graphics, grid, lattice, ggplot2), and is extensible through CRAN and Bioconductor packages; it runs on Windows, macOS and Linux and integrates with tools such as RStudio, Jupyter (IRkernel), Rcpp, reticulate and sparklyr. R is distributed under the GNU General Public License.
Firecracker is an open-source virtualization technology developed at Amazon Web Services for running secure, multi-tenant container and function workloads in lightweight virtual machines called microVMs. Its virtual machine monitor uses Linux KVM and a minimalist device set to combine hardware-virtualization isolation with container-like startup speed and resource efficiency. Firecracker exposes a host API for configuring and managing microVMs and is designed for serverless workloads and isolated environments, including agent environments. It is distributed under the Apache License 2.0 and can be built from source or obtained as release binaries.
OverlayBD is a tool for on-demand loading of environment/container images that fetches image data lazily and caches retrieved chunks to local disk so an environment can start before the entire image is downloaded. It is presented in videos as a lazy-image loader using a local disk cache to reduce upfront image pull time.
QuickJS is a JavaScript engine developed by Fabrice Bellard. In the cited context, scriptc uses it to provide dynamic-mode execution for npm dependencies. Its main documentation is provided in HTML and PDF form in the project repository.
LanguageTool is open-source proofreading software that checks spelling, grammar, and style in English, Spanish, French, German, Portuguese, Polish, Dutch, and more than 20 other languages. It detects errors that basic spell checkers may miss and can be used through a graphical interface, command line, server, HTTP API, or Java integration. The core repository is available under the LGPL 2.1 or later.
QEMU is an open-source machine emulator and virtualizer (the QEMU Project) that performs system and user-mode emulation across many CPU architectures (x86, ARM, PowerPC, RISC-V, etc.) and boots disk images or ISO files as guest systems. It provides a broad set of emulated peripherals and virtual devices — including virtual GPU devices (virtio-gpu, qxl, cirrus, and related 3D paths such as virgl) — supports hardware-assisted virtualization (KVM) and device passthrough (VFIO), and is distributed under the GNU General Public License.
WebGL 2.0 is a web graphics API standardized by the Khronos Group that provides GPU-accelerated 2D and 3D rendering inside web browsers. It is based on OpenGL ES 3.0 and exposes features including GLSL ES 3.00 shaders, vertex array objects, uniform buffer objects, transform feedback, multiple render targets, 3D textures and floating‑point texture formats; it renders to HTMLCanvasElement framebuffers via a WebGL2RenderingContext. WebGL 2.0 is implemented in modern browsers and is used from JavaScript and WebAssembly for in‑browser rendering and graphics applications.
Pluggy is a Brazilian open-finance infrastructure provider offering commercial APIs to access bank account and transaction data and to initiate payments and transfers across Brazilian banks. It connects to institutions in Brazil's Open Finance ecosystem and is distributed as a SaaS API used by fintechs and financial applications.
Claude Desktop is Anthropic's desktop application for interacting with Claude, its AI assistant. It can host extensions such as Token-Saver, which sends selected passages from a PDF instead of the full document.
An AI API service from MiniMax that provides access to generation capabilities and context preprocessing associated with the MiniMax-H3 language model. The API includes a 2K generation stage and a context preprocessor that are not included in MiniMax-H3's open-weight release.
Moonshine Voice is an open-source AI toolkit for developers building real-time voice agents and applications. It provides on-device speech-to-text, intent recognition, and text-to-speech, including streaming transcription that processes audio while the user is still speaking. The project offers speech-to-text models ranging from higher-accuracy models to approximately 1 MB models and provides one library for Python, JavaScript/WASM, iOS, Android, macOS, Linux, Windows, and Raspberry Pi. It is used by Gemma Translator for speech recognition and speech output, and can run without an account or API keys. The project is distributed under the MIT License, including its models by default, with legacy non-streaming models for non-English languages covered by a non-commercial Moonshine Community License.
Ratatui is a Rust crate for building text-based terminal user interfaces (TUIs), including command-line applications, dashboards, and interactive console programs. It provides terminal initialization and restoration, frame-based rendering, widgets, event handling, templates, and example applications; the videos identify it as the Rust terminal UI used by the WP2I WhatsApp client.
whatsmeow is a Go library for the WhatsApp Web multidevice API, used as a library bridge by the WP2I WhatsApp client. It supports sending and receiving text and media messages in private chats and groups, group management, invite links, typing notifications, delivery and read receipts, app-state synchronization, retry receipts, and experimental status messages. Calls and broadcast-list messages are not implemented.
Chafa is a command-line utility and C library for converting image data, including animated GIFs, into terminal graphics such as ANSI/Unicode character art. It supports output for a range of devices and terminal emulators, including terminals using formats such as Sixel, Kitty graphics, and iTerm2. The library provides a documented public API, while the command-line frontend can be built from source or installed through distribution packages. Both are licensed under LGPLv3 or later.
Spring Boot is a Java framework from the Spring project for creating stand-alone, production-grade Spring applications and services. It supports executable Java applications launched with `java -jar`, traditional WAR deployments, and a command-line tool for running Spring scripts. Its conventions provide common features such as embedded servers, security, metrics, health checks, and externalized configuration, without code generation or mandatory XML configuration. In the cited video, Spring Boot is the target project environment for the Lithe-IDEA Java IDE.
A build automation and project management tool for Java and JVM-based projects, developed under the Apache Software Foundation. Maven uses an XML Project Object Model (POM) to declare dependencies and build configuration, implements a lifecycle with phases (compile, test, package, install, deploy), resolves artifacts from Maven Central (and other repositories), and produces build artifacts such as JARs and WARs; it is extensible via plugins and integrates with IDEs (IntelliJ IDEA, Eclipse) and CI systems. Maven is distributed under the Apache License 2.0 and runs on the Java platform (JDK).
Eclipse JDT Language Server (eclipse.jdt.ls) is the Eclipse Foundation's Java language server that implements the Language Server Protocol to provide Java editing features such as semantic navigation (go-to-definition), code completion, diagnostics, refactorings, formatting, and code actions. It is maintained by the Eclipse JDT project, distributed under the Eclipse Public License, runs on the JVM, and can be used with any LSP-compatible editor or IDE.
Shortcuts is Apple's automation app for creating and running workflows of native and third-party app actions on iPhone, iPad, Mac and Apple Watch. Developed and distributed by Apple and bundled with iOS, iPadOS and macOS, it composes sequences of actions with variables, parameters and scripting blocks (conditionals, loops), supports web requests, file and media operations, and integrates with Siri, widgets, HomeKit and automation triggers; shortcuts can also be invoked from other apps via a URL scheme and pass data between apps.
Cloudflare Tunnel (formerly Argo Tunnel) is a secure tunneling service from Cloudflare that exposes local HTTP and TCP services to the Cloudflare edge by creating an outbound connection from a host to Cloudflare's network. It is implemented by the open-source cloudflared client/daemon, supports named tunnels and ingress rules, and integrates with Cloudflare Zero Trust / Access for authenticated routing and access control. The service is distributed by Cloudflare and the cloudflared client is published on Cloudflare's GitHub.
A text-recognition API in Apple's Vision framework that detects and recognizes text in images and video and returns recognized strings, confidence scores, and VNTextObservation bounding boxes (normalized coordinates). Provided by Apple as part of the Vision framework (API types include VNRecognizeTextRequest and VNTextObservation), it runs on iOS, iPadOS, macOS and visionOS and is distributed to developers via Xcode and the Apple Developer Documentation; outputs can be converted to screen/tap coordinates for UI automation or accessibility uses.
M5Stack is a modular hardware and open-source software platform for IoT prototyping and deployment. Its standard 5 × 5 cm system uses stackable functional modules built around an ESP32 microcontroller, with features including a microSD card slot, USB-C port, and expansion connectors. M5Stack also provides multifunctional base modules and ODM services for projects ranging from prototypes to mass production. In the cited video, M5Stack hardware is used by VibeWatch as a smartwatch interface for controlling multiple AI coding agents.
Skia is a complete 2D graphics library for drawing text, geometries, and images. It is a rendering system that the referenced React Native theme-transition library deliberately avoids.
arXiv is an open-access repository of scholarly preprints (e-prints) operated by Cornell University Library. It hosts papers across physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering and economics, assigns persistent arXiv identifiers, provides full-text (PDF/TeX) and metadata, and exposes content via its website, OAI-PMH and a public API; submissions are subject to basic moderation and some categories require endorsement. The service is free to read and is funded and governed through Cornell University Library together with a membership program and institutional/foundation supporters (including the Simons Foundation).
Open CASCADE Technology (OCCT) is an open-source software development platform and CAD geometry kernel for 3D surface and solid modeling, CAD data exchange, and visualization. Its functionality is provided mainly as C++ libraries for applications in computer-aided design, manufacturing, and engineering analysis; the videos describe it as compiled to WebAssembly for browser-based parametric CAD. OCCT is distributed under the GNU Lesser General Public License version 2.1 with a special exception, and may alternatively be used under an Open CASCADE commercial license or contractual agreement.
Replicad is a browser-based CAD library that provides a programmatic API for building 3D models with code. Videos describe it as a browser-facing CAD layer built on Open Cascade, indicating it leverages Open Cascade for core geometry and runs client-side.
ROS 2 is an open-source robotics middleware (the successor to ROS/Robot Operating System) developed by Open Robotics and the ROS community that provides publish/subscribe topics, services, actions, parameters and lifecycle management for distributed robot software. It uses the DDS (Data Distribution Service) transport layer with implementations such as eProsima Fast DDS, Eclipse Cyclone DDS and RTI Connext, and includes client libraries (rclcpp, rclpy), tooling (colcon) and rosbag2 for logging and replay of topics. ROS 2 runs on Linux, Windows and macOS, integrates with simulators and robot stacks, and is distributed under the Apache 2.0 license.
Rosbag2 is a ROS 2 tool for recording and playing back timestamped messages exchanged by robotics systems. It records all available topics or selected topics, can discover topics that appear during recording, and stores the data in a bag directory for later inspection or replay. It is available through the ROS 2 command-line interface as `ros2 bag`, with commands for recording, playback, conversion, information, listing, bursting, and reindexing; its recorder and player can also run as `rclcpp` components.
MCAP (pronounced "em-cap") is an open-source container file format for multimodal log data that records multiple channels of timestamped, pre-serialized messages. It is designed for replayable recordings and analysis in pub/sub and robotics applications, and serves as a storage format for message streams and telemetry. The format focuses on efficient, timestamped multimodal logging suitable for replay and tooling integration.
Nmap, the Network Mapper, is a network-discovery and port-scanning tool for identifying devices, open ports, operating systems, and services. It can scan targets such as a home-lab network and includes Zenmap as a graphical front end. The project provides source code and binary installers for Windows, macOS, and Linux, and is released under a custom license based on but incompatible with GPLv2; commercial redistribution licensing is also available.
Zigbee2MQTT is an open-source Zigbee-to-MQTT bridge developed in the Koenkk project, allowing Zigbee devices to operate without their vendor-specific bridge or gateway. It uses zigbee-herdsman to communicate with a hardware adapter, zigbee-herdsman-converters to map device models to supported Zigbee clusters, and the Zigbee2MQTT module to translate Zigbee messages into MQTT messages while tracking device state in a JSON database. It provides web interfaces for monitoring and configuration and can integrate with home-automation systems such as Home Assistant, Homey, Domoticz, Gladys Assistant, and ioBroker.
A task and reminders app developed by Apple Inc., bundled with iOS, iPadOS, macOS, watchOS and available via iCloud.com. It stores and presents reminder items (titles, dates/times, location-based triggers, subtasks, notes, attachments and tags) as lists and issues local and push notifications; it normally syncs via iCloud but also supports CalDAV synchronization with third-party CalDAV servers (the videos cite compatibility with a service called Two Due Duty). The app is proprietary software distributed as part of Apple operating systems and tied to an Apple ID for iCloud sync.
Nextcloud is an open-source content collaboration platform used by organizations to manage and share content. In the stated context, it acts as a supported CalDAV synchronization client for Two Due Duty.
An open-source animation library for JavaScript, React, and Vue. It combines JavaScript with native browser APIs for GPU-accelerated animations and provides APIs for gestures, springs, layout transitions, scroll-linked effects, and timelines. The React package is imported from `motion/react`, the JavaScript package from `motion`, and the Vue package from `motion-v`; Motion is MIT licensed. The video identifies it as the animation foundation for Amicro.
Lucide is an open-source icon library maintained by the community and forked from Feather Icons. It provides more than 1,600 vector SVG icons for digital and physical projects, with official packages for React, Vue, Svelte, Solid, Preact, React Native, Angular, Astro, and static use, plus a Figma plugin. The library is free for commercial and personal use under the ISC License and does not accept brand logos.
Krea is a web-based AI creative suite for generating, editing, and enhancing images, video, and 3D assets. Its image tools accept text and image prompts and include prompt management, galleries, and team or shared workspaces for organizing and collaborating on generative assets. It is offered as a SaaS platform at krea.ai with free and paid tiers.
Fathom is a commercial AI meeting assistant and call recorder that captures video calls, including Zoom meetings, and generates searchable transcripts, highlights, shareable clips, summaries, and notes. It is provided as a web app with Zoom and other conferencing-platform integrations, and supports exporting meeting artifacts for sharing and archival.
Steno is an open-source, privacy-first desktop AI notepad and meeting notetaker for Windows and macOS, intended for confidential conversations. It records microphone and system audio, provides live or post-meeting transcription with speaker labels, generates summaries, and supports natural-language queries across individual notes or the meeting library. Its local mode uses on-device AI models; on Apple Silicon, live transcription uses Parakeet TDT v3 through MLX, while Whisper supports post-stop transcription in 99 languages. On macOS 14.4 and later, native Core Audio Tap captures both sides of virtual meetings without a virtual audio cable and with selectable microphone input. Steno can detect meeting start and end events, offer to start or summarize recordings, append recordings to existing notes, provide global recording shortcuts, and include user notes in generated summaries. Notes, transcripts, and summaries are stored as editable Markdown, with one-way synchronization to an Obsidian vault and report templates for structured outputs. The project states that local recordings, transcripts, and summaries remain on the device and have no cloud usage limits; users can alternatively configure hosted models through OpenAI, Anthropic, AWS Bedrock, or a custom API endpoint.
Ailin is an open-source, self-hosted collective-intelligence engine and OpenAI-compatible interface in which AI models collaborate inside a collective model rather than being routed to a single model. It indexes models from frontier APIs, open-weight systems, and its own model family, coordinating them through dozens of strategies such as debate, critique, and synthesis. The platform records coordination decisions and decision provenance for each request, can return reasoning and cost information with an audit trail, and includes model discovery, an audit layer, and a closed-loop training pipeline; some components are still maturing.
Fractera is an open-source, self-hosted agent-engineering infrastructure platform from Fractera. Given an Ubuntu 24.04 VPS, its installer configures the operating system environment, Nginx routing, HTTPS certificates, role-based authentication, a database, local object storage, browser terminals, and a starter application or other public Git repository. The resulting deployment keeps code, data, and agent execution on the user’s server rather than relying on a managed cloud platform. Its deployment architecture is described as deterministic and MCP-first. The server includes a browser-accessible multi-agent development environment coordinated by the Hermes orchestrator, with five specialized code-generation engines and shared local RAG memory; the platform is also described as supporting agents for coding, marketing, and sales. Fractera can build the infrastructure around multiple application frameworks and Git repositories, rather than requiring a single frontend stack. The current Next.js starter is an approximately 50,000-line enterprise boilerplate with multilingual routing, static SEO trees, a SQLite write-ahead-logging database, NextAuth v5 session state, Next.js parallel routing, and on-demand incremental static regeneration. Its largely static base architecture is intended to limit agents’ repeated scanning and reconstruction of workspace context; the project claims its optimizer can reduce token costs by up to 90% by limiting context-window inflation.
An open-source marketing-agent team from Vercel Labs built on the eve framework. Users submit launch planning, copywriting, or SEO requests through Slack or the eve terminal TUI; a lead agent loads the shared brand-context document, writes a self-contained brief, routes it to one of five specialists—product marketing, content marketing, social media, SEO, or email—and returns the resulting deliverable. The lead does not write deliverables itself, and specialists work independently without shared conversation history; the product marketer alone maintains the brand-context document, which the other specialists read at the start of a task. Newsletter work passes from the content marketer to the email specialist for inbox adaptation and sending. Outputs are delivered through Notion, Typefully, Resend, or stored audit artifacts. The system pauses irreversible operations—including Resend sends and deletes, Typefully deletes and scheduled publishing, and Notion page moves—for approval in Slack or the terminal; its email specialist is restricted to an allowlist of Resend tools. Deployment provisions Notion, Resend, and Slack connectors and a Vercel Blob store, and requests a Typefully API key. The implementation uses TypeScript, Vercel Connect, MCP connections for Notion, Typefully, and Resend, Vercel AI Gateway for model access, and Vercel Sandbox for reference files and shell commands. Email campaigns require a verified Resend sending domain and at least one segment configured in Resend.
NVIDIA-labs Object Oriented Agents (NOOA) is a model-agnostic Python framework for building AI agents as ordinary Python objects. A single agent class combines typed state, capabilities, prompts, and interfaces: fields represent state, methods represent capabilities, docstrings provide prompts, and type annotations define contracts. Methods with an ellipsis body become LLM-driven agentic loops, while methods with ordinary bodies remain deterministic Python. The runtime lets a model act by writing Python in a Jupyter-style REPL with access to the agent object, imports, and helper functions. It supports typed inputs and outputs with automatic retries, live-object arguments passed by reference, and model-callable context and event APIs, and can use hosted or local models through LiteLLM-supported providers. The core framework is distributed as the `nooa` Python package through PyPI or the project repository. Separate packages or extras provide a CLI and trace viewer, Agent Client Protocol support, long-term memory, benchmarking, and evaluation tooling. NOOA is research software whose agents may execute LLM-generated code; its AST checks and module deny-lists are defense-in-depth guardrails rather than a containment boundary, so the documentation recommends running such agents in an OS-level sandbox such as a container, virtual machine, or NVIDIA OpenShell.