AI Agents & Automation

GPTFuzz is an open-source tool designed for red teaming large language models (LLMs) by automatically generating jailbreak prompts. This process helps identify vulnerabilities and weaknesses in AI models, ultimately enhancing their robustness and security. The repository provides the official codebase for "GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts." It includes datasets for harmful questions and human-written templates, along with a finetuned RoBERTa-large model for judgment. Researchers can use GPTFuzz to generate their own adversarial templates and contribute to building a general black-box fuzzing framework for LLMs.

mlx-lm is a Python package designed for generating text and fine-tuning large language models (LLMs) specifically on Apple silicon using the MLX framework. It offers seamless integration with the Hugging Face Hub, allowing users to easily access and utilize a vast array of LLMs with simple commands. Key features include support for quantizing models, uploading them to the Hugging Face Hub, and performing both low-rank and full model fine-tuning, even with quantized models. The package also provides distributed inference and fine-tuning capabilities with `mx.distributed`, and tools for efficient handling of long prompts and generations through a rotating fixed-size key-value cache and prompt caching.

neurojs is an open-source JavaScript framework designed for deep learning and reinforcement learning applications within the browser environment. While it mainly focuses on reinforcement learning, it is versatile enough for various neural network-based tasks. The library includes practical examples and demos, such as a 2D self-driving car visualization, to showcase its capabilities. It supports advanced features like uniform and prioritized replay buffers, advantage-learning, and models such as deep-q-networks and actor-critic (via deep-deterministic-policy-gradients). neurojs also allows for binary import and export of network configurations, including weights, and is built for high performance. However, development on neurojs is no longer actively maintained, with the recommendation to use more general frameworks like TensorFlow-JS.

hum.ai is dedicated to building advanced multimodal foundation models designed for practical, real-world applications. Their core focus is on leveraging satellite remote sensing and ground truth data to train these models, aiming to develop Artificial General Intelligence (AGI) for a deeper understanding of the natural world. The technology developed by hum.ai is currently being utilized in critical sectors such as nature conservation, carbon dioxide removal initiatives, and by various government agencies. This positions hum.ai at the forefront of applying AI to solve complex environmental and scientific challenges, providing robust solutions for data analysis and predictive modeling in these domains.

Glow-TTS is an open-source generative flow model designed for text-to-speech (TTS) synthesis, utilizing a monotonic alignment search. Unlike many parallel TTS models, Glow-TTS does not require external aligners, making it a self-contained solution for generating mel-spectrograms from text. By combining the properties of flows and dynamic programming, it efficiently searches for the most probable monotonic alignment between text and the latent representation of speech. This approach ensures robust TTS, capable of generalizing to long utterances, and enables fast, diverse, and controllable speech synthesis. The model achieves significant speed-up over autoregressive models like Tacotron 2 with comparable speech quality and can be extended to multi-speaker settings. It also supports integration with vocoders like HiFi-GAN for improved synthesis quality.

External-Attention-pytorch is a comprehensive GitHub repository offering PyTorch implementations of numerous attention mechanisms, Multi-Layer Perceptrons (MLPs), re-parameterization techniques, and convolution operations. This resource is designed for developers and researchers looking to deepen their understanding of these fundamental components in deep learning models. It includes detailed examples and usage instructions for over 30 different attention mechanisms, such as External Attention, Self Attention, MobileViT Attention, and many more. Additionally, it covers various backbone architectures like ResNet and MobileViT, several MLP types, and re-parameterization methods like RepVGG. The repository serves as a valuable educational and practical toolkit for implementing advanced neural network architectures.

Playgent offers specialized reinforcement learning (RL) environments tailored for the finance and banking sectors. These environments simulate realistic market conditions, document processing workflows, and complex decision-making scenarios, mirroring real-world trading, compliance, and operational tasks. The platform provides challenging financial tasks, verification rubrics, and production-ready environments for post-training AI agents. Playgent emphasizes that the quality of the environment directly impacts the quality of the agent, and their expert-curated tasks are benchmarked to ensure high performance. Examples include environments for LBO returns analysis, earnings normalization, and M&A synergy analysis, all designed to help agents excel at financial decision-making.

Parlant is an open-source interaction control harness designed for customer-facing AI agents, optimizing for controlled, consistent, and predictable customer interactions with Large Language Models (LLMs). It streamlines the development and maintenance of enterprise-grade B2C and sensitive B2B interactions, ensuring they are compliant and on-brand. Parlant addresses the challenges of conversational context engineering by providing an agentic harness that optimizes context engineering for conversational use cases. It allows developers to define rules, knowledge, and tools once, with the engine dynamically narrowing the context in real-time to what's immediately relevant for each turn of the conversation. This approach ensures maximum control over conversation experience, prevents unwanted behaviors by applying constraints, and offers a rapid feedback loop for product adjustments.

EverMemOS is a memory operating system designed to equip AI agents with persistent, proactive, and self-evolving memory. It addresses the limitations of stateless LLMs by enabling them to maintain context across days, sessions, and platforms, effectively turning them into intelligent agents that can truly remember. The tool offers multimodal retrieval and ingestion (mRAG), allowing it to parse and store various data types like PDFs, images, spreadsheets, and URLs through a single API. EverMemOS records agent trajectories as 'Cases,' distills patterns into reusable 'Skills,' and facilitates learning from past experiences. It also features a 'Memory Bank' interface for transparent management of user, group, and agent memories, including the ability to inspect and edit generated Skills. EverMemOS is available as a cloud service or an open-source solution.

qomplement is an AI agent designed to automate desktop tasks across various software applications, significantly streamlining workflows by automating repetitive processes. This tool is particularly useful for tasks such as document filling and data entry, enhancing overall productivity for individuals and businesses. By leveraging AI, qomplement aims to reduce manual effort and potential errors associated with routine administrative work. Its focus on automating desktop interactions makes it a valuable asset for improving efficiency in daily operations.

Swarmer offers combat-proven collaborative autonomy software designed to enable a single operator to command hundreds of drones across various domains. The platform emphasizes AI-driven navigation and scalable autonomy, streamlining missions with advanced capabilities. Key features include encrypted over-the-air updates to align with real-world data, autonomous navigation, real-time combat data processing, AI-driven collaboration, an intuitive user interface, and a modular software stack. Swarmer's technology is trusted by military units and focuses on maintaining capabilities aligned with frontline tactics. The software is built for demanding environments, ensuring robust and efficient drone operations.

Wolfe by Slideworks is an AI-powered management consultant designed to assist with a wide range of business questions and challenges. It leverages advanced generative language models and the expertise of top-tier management consultants to provide strategic guidance. Wolfe can act as a co-pilot for tasks such as research, drafting, analysis, and communication, making these processes more efficient. It helps users create presentation storylines, develop frameworks for projects like digital transformation, solve business problems, optimize pricing, and analyze data for insights. Founded by ex-consultants and developers in partnership with Slideworks, Wolfe aims to augment corporate teams and consultants with cutting-edge AI capabilities.

Curebase is an AI-native eClinical platform designed to unify sponsors and sites on a single system, accelerating clinical trials from study startup to database lock. It provides a comprehensive suite of tools including ePRO/eCOA for patient-reported outcomes, eConsent for electronic informed consent, Electronic Data Capture (EDC), and robust patient recruitment capabilities. The platform also features dedicated site software (Sitebase) to streamline patient management and automate workflows for research sites. Curebase aims to improve data quality and boost participant engagement, adapting to the needs of biotech, MedTech, pharma, and CROs, making it suitable for lean teams and global programs alike.

claude-code-hooks-multi-agent-observability offers a comprehensive system for real-time monitoring and visualization of Claude Code agents, particularly useful for multi-agent orchestration with Claude Opus 4.6. By tracking hook events, the system provides deep observability into agent behavior, including tool calls, task handoffs, and agent lifecycle events. It features a robust architecture that captures, stores, and visualizes events in real-time, supporting multiple concurrent agents with session tracking, event filtering, and live updates. The system includes a Bun-powered TypeScript server for event processing and a Vue 3 client for interactive visualization, complete with a dual-color design, multi-criteria filtering, and a live pulse chart. Developers can easily integrate the observability hooks into their projects to gain insights into their Claude Code agent operations.

HPSv2 is a comprehensive benchmark designed for evaluating human preferences in text-to-image synthesis. It features the Human Preference Dataset v2 (HPD v2), a large-scale dataset comprising 798k preference choices across 430k images, and the Human Preference Score v2 (HPS v2), a preference prediction model trained on HPD v2. This tool allows users to compare images generated with the same prompt and provides a fair, stable, and easy-to-use set of evaluation prompts. It supports benchmarking models across various styles like Animation, Concept-art, Painting, and Photo, and offers functionalities for custom model evaluation and preference model assessment.

JustAHuman offers a unique gamified platform for 3D asset evaluation and labeling, allowing users to earn rewards while contributing to data annotation. Players accumulate points by completing challenges, which can then be converted into game credits, GenAI service provider credits, or crypto. This innovative approach aims to improve the efficiency and accuracy of AI model training by engaging users in a fun and rewarding way. The platform is designed to connect game creators with a community that can help process and label their 3D assets, making it a valuable resource for both players and developers.

ChatGPT Easy Folders, also known as ChatGPT Toolbox, is a Chrome extension designed to significantly enhance the ChatGPT user experience by providing robust organization and management features. It allows users to organize their ChatGPT conversations into unlimited folders and subfolders, pin important chats, and conduct advanced searches across their entire chat history. Beyond organization, the tool offers bulk export capabilities in TXT and JSON formats, a media gallery for DALL-E images, and custom themes. For power users, it includes a prompt library with expert prompts and a prompt chaining feature for workflow automation. It supports RTL languages and offers cross-device sync with its paid plans, making it a comprehensive solution for managing ChatGPT interactions.

ChatGPT-clone provides an enhanced interface for interacting with ChatGPT, focusing on a better user experience. This open-source project allows for local deployment and customization, making it suitable for developers and technical users who want more control over their AI chat environment. Key features include the ability to configure an OpenAI API key and base URL, supporting reverse proxies for queries. While development was temporarily halted, it is actively seeking contributions to further improve functionalities such as conversation management, user preferences, theme changing, and speech output/input integration. It's built with Python, JavaScript, CSS, and HTML, offering a flexible foundation for further development.

JourneoAI is an AI-powered travel planning platform designed to create personalized itineraries quickly and efficiently. Users can input their budget, desired vibe (e.g., romantic, adventure, foodie), and destination to receive a full trip plan in seconds. The tool provides daily lineups of restaurants, bars, hidden spots, and activities, all balanced for time and budget. JourneoAI also offers real-time flight and hotel suggestions from trusted partners like Expedia and Travelpayouts, allowing users to book directly. It supports quick regeneration of itineraries if plans change and can adapt to dietary needs, allergies, or mobility requirements. The platform works directly in the browser, eliminating the need for app downloads, and caters to solo, couples, or family travel styles.

Evalyze is an AI-powered platform designed to streamline the fundraising process for startups. It intelligently matches founders with suitable VCs and angel investors based on stage, sector, and growth, significantly reducing the time spent on manual outreach. Beyond matching, Evalyze offers a sophisticated pitch deck analysis engine that provides actionable feedback and an Investor Readiness Score, helping founders understand how their deck will be perceived by investors. This comprehensive assessment helps refine their story and strengthen their presentation. The platform aims to automate various fundraising tasks, with upcoming features like automated investor messaging and an AI fundraising CRM, allowing founders to focus more on closing deals and less on administrative overhead. It's trusted by thousands of founders for its efficiency and insightful analytics.

Socrates is an advanced AI tool designed for comprehensive document analysis, enabling users to unlock complete and accurate answers from PDFs, DOCXs, EPUBs, and text files. Its standout "Deep Dive" feature intelligently breaks down lengthy documents, creating multiple search indexes for thorough analysis. Users can also build custom AI workflows with "Flow AI" and compare multiple documents using "Table AI." A key differentiator is its support for local LLMs, allowing for document analysis without sending data to the cloud, which is ideal for users prioritizing data privacy and security. Socrates also offers the ability to ask questions across multiple documents, search specific pages, and save frequently used prompts, making it a versatile solution for researchers and professionals.

StudyNinja is an AI-enhanced study buddy designed to elevate the academic journey for students. It integrates technology with intuitive design to boost efficiency and productivity, offering features like personalized study goal crafting and progress monitoring. The platform excels in assignment management, allowing users to create and collaborate on documents, and provides robust to-do lists for task management. Note-taking is redefined with options to share insights and export notes as PDF or Word. A standout feature is the Curious AI Tutor, ready to assist with questions and broaden knowledge through insightful conversations. StudyNinja also streamlines group projects with dynamic collaboration tools, enabling delegation and effective communication. It supports active learning with interactive tools like digital flashcards and innovative note-taking, all accessible on PC, tablet, or mobile without software installations or updates.

cog-stable-diffusion provides an implementation of the Diffusers Stable Diffusion v2.1 model, packaged as a Cog model. This approach allows machine learning models to be distributed and run as standard containers, simplifying deployment and ensuring consistent environments. Users can download pre-trained weights and then execute predictions by providing prompts, enabling the generation of images. This tool is particularly useful for developers and researchers who need to integrate Stable Diffusion capabilities into their applications or workflows, offering a streamlined way to manage and deploy the model.

open-llms is a comprehensive GitHub repository that serves as a curated list of open Large Language Models (LLMs) explicitly licensed for commercial use, including Apache 2.0, MIT, and OpenRAIL-M. This resource is invaluable for developers, researchers, and businesses looking to integrate open-source LLMs into their applications without licensing concerns. The repository details each model's release date, available checkpoints, associated research papers or blog posts, parameter sizes, context lengths, and specific licenses. It also includes a dedicated section for open LLMs tailored for code generation, offering insights into models like SantaCoder, CodeGen2, and StarCoder. Contributions to the list are welcomed, ensuring it remains up-to-date with the latest commercially viable open LLM releases.