AI Agents & Automation

Free Agent Skill Builder is a powerful AI tool designed to simplify the creation of professional AI agent skills. Users can easily describe the type of skill they need and drag-and-drop reference files, templates, or scripts to provide context. The tool automatically handles complex aspects like structure, token optimization, and adherence to best practices, eliminating the need for manual specification or schema writing. It supports text files and PDFs for context, generating a production-ready skill package in seconds. These skills can be used with Maia, Claude AI, Claude Code, Gemini CLI, and OpenAI Codex, making it a versatile solution for developers and non-technical users alike. The tool is free and provided by ModularMind, ensuring users retain full ownership of their data and generated skills.

RD-Agent is an innovative AI tool developed by Microsoft, designed to automate critical and high-value aspects of industrial research and development (R&D). Focusing primarily on data-driven scenarios, it streamlines the development of models and data through a unique framework that includes 'R' for proposing new ideas and 'D' for implementing them. The tool has demonstrated leading performance as a machine learning engineering agent on MLE-bench, a comprehensive benchmark for AI agents. Additionally, RD-Agent offers a data-centric, multi-agent framework for quantitative finance, automating the full-stack R&D of quantitative strategies. It supports various applications, including an automatic quant factory, data mining, research copilot, and Kaggle agent functionalities, continuously expanding its methods and scenarios to boost productivity.

Apache Burr is an open-source Python framework designed for building applications that make decisions, such as chatbots, AI agents, and simulations. It allows developers to express applications as state machines, managing state, tracking complex decisions, and integrating human feedback. Burr includes a UI for real-time monitoring, tracing, and introspection of execution telemetry, along with pluggable persisters for saving and loading application state. It is framework-agnostic and supports both LLM and non-LLM use cases, offering integrations for various systems and the ability to build custom actions. The tool emphasizes tying together models, APIs, and data management in a scalable and transparent way.

BrowserWing is an open-source platform designed to empower AI agents with efficient and reliable browser control. It translates browser actions into MCP commands or Claude Skills, enabling seamless integration with various AI tools. The platform boasts over 26 HTTP API endpoints for comprehensive browser automation and includes a built-in AI agent for conversational tasks. Key features include visual script recording, flexible export options for AI tool integration, and intelligent data extraction powered by LLMs like OpenAI and Claude. BrowserWing supports robust session management, making it ideal for AI-driven tasks, browser automation, and boosting productivity.

The Self-hosted AI Starter Kit is an open-source Docker Compose template designed to swiftly initialize a comprehensive local AI and low-code development environment. Curated by n8n, it combines the self-hosted n8n platform with a curated list of compatible AI products and components. This kit allows users to build secure, self-hosted AI workflows, including AI agents for scheduling appointments, summarizing company PDFs securely, creating smarter Slack Bots, and analyzing private financial documents at minimal cost. It includes n8n for low-code automation, Ollama for running local LLMs, Qdrant for vector storage, and PostgreSQL for data handling. The kit supports Nvidia, AMD, and CPU setups, with specific instructions for Mac users.

SimCSE is an open-source framework that implements simple contrastive learning for generating high-quality sentence embeddings. It supports both unsupervised learning, using only standard dropout as noise, and supervised learning, incorporating annotated pairs from NLI datasets. The framework offers pre-trained models, including BERT-based and RoBERTa-based versions, and provides an easy-to-use Python package for encoding sentences and computing similarities. Users can also integrate SimCSE models with HuggingFace's Transformers library. The repository includes comprehensive training and evaluation code, allowing researchers and developers to reproduce results or train custom models on their own datasets. It is particularly useful for tasks requiring robust semantic textual similarity and downstream transfer capabilities.

Physical Intelligence is at the forefront of bringing general-purpose AI into the physical realm. The company comprises engineers, scientists, and roboticists dedicated to developing advanced learning algorithms. Their primary goal is to create a foundational model capable of controlling any robot to perform a wide array of tasks. This involves research into steerable robotic foundation models, efficient online reinforcement learning for precise manipulation, and multi-scale embodied memory for complex, long-duration tasks. They are also exploring real-time action chunking and vision-language-action models that train quickly, generalize well, and learn from experience, aiming to enable a "Cambrian explosion" of robotics applications.

Imbue is dedicated to empowering humans in the age of AI by building products that give individuals control over their digital world. Their flagship product, Sculptor, is a user interface designed for coding agents, aiming to make AI coding reliable, collaborative, and accessible to everyone. Sculptor allows users to run multiple coding agents safely in containers and instantly test changes. Imbue's mission extends to making software creation accessible to all, enabling every person to shape their computational future. They also conduct research, including training large language models, to advance AI capabilities.

Syncortex is a technology company specializing in AI-driven digital transformation, offering a suite of services designed to help businesses automate, adapt, and act using advanced AI. Their core offerings include Agentic AI solutions that design intelligent agents for autonomous task execution and workflow orchestration, AI-Accelerated Coding to build products faster and more cost-effectively by combining AI-powered tools with expert engineering, and Data Engineering services for building robust data pipelines and real-time analytics. Syncortex also provides Analytics services to transform data into actionable intelligence through modern BI tools and predictive insights. They emphasize building agentic systems that are not just intelligent but human-centered, focusing on clarity, responsibility, and trust.

Build by Nvidia offers NVIDIA NIM APIs, a comprehensive platform designed for building enterprise generative AI applications. It provides access to a wide range of leading AI models, optimized for efficient inference. Users can obtain free serverless APIs for development, with options for accelerated deployment via DGX Cloud or self-hosting on their own GPU infrastructure. The platform includes continuous vulnerability fixes and offers step-by-step playbooks, such as setting up NemoClaw for secure personal AI agents. Additionally, it features NeMo Data Designer for creating high-quality, domain-specific synthetic datasets at scale, and blueprints for building AI agents, video search and summarization tools, and data flywheels for continuous optimization.

Salesboom is an award-winning cloud CRM software that has been providing customizable, affordable, and integrated solutions since 2003. It unifies sales, marketing, and customer service on a single user-friendly platform, helping businesses manage customer interactions, automate workflows, and gain AI-powered insights. The platform includes features like Sales Force Automation (SFA), lead and contact management, marketing automation, and comprehensive support tools. Salesboom offers a Fast Track Implementation Program, aiming for deployments within 30-40 days, and integrates with various enterprise systems including ERP, accounting, and popular communication tools. It is designed to be mobile-friendly, allowing businesses to operate on the go.

Chatbot GPT3 Plugin is a versatile tool built on the Bubble platform, designed for creating web applications without requiring any code. It offers a unique approach by allowing users to seamlessly switch between AI prompting and visual editing interfaces, catering to different development preferences and skill levels. The platform hosts all applications on its cloud infrastructure, simplifying deployment and management. This tool is ideal for individuals and businesses looking to rapidly develop and launch web applications, leveraging AI capabilities for enhanced functionality and efficiency in the development process.

TigerBot is a multi-language, multi-task Large Language Model (LLM) designed to serve as a foundational platform for developing custom LLMs. It stands out for its strong focus on Chinese language innovation, featuring the world's first Chinese clinical terminology system (MedCT) and corresponding medical NER models, which have achieved new SOTA results in both English and Chinese medical NER tasks. The tool also provides real clinical annotation data. TigerBot supports extensive context lengths, up to 100K for its 70b and 13b chat models, and offers function calling capabilities. It is open-source and available on platforms like Hugging Face and ModelScope, making it accessible for research and commercial use. The platform has undergone continuous upgrades, improving its comprehensive capabilities across various models and offering features like search-augmented modes and document enhancement (RAG).

YC-Killer is an open-source library of enterprise-grade AI agents developed by Singularity Research, aiming to democratize artificial intelligence by providing free, production-ready alternatives to overvalued commercial startups. The project features a suite of sophisticated AI agents, including an Agentic Deep Research tool for comprehensive analysis, an Agentic Quant Hedge Fund for autonomous trading, and Jarvis, an Agentic Executive Agent for personal assistance. Other agents include an Agentic Call Center, an Agentic AI Hospital for medical care, an Agentic Professor for tutoring, and an Agentic Accounting Firm. These agents are built with robust technologies like TypeScript/Node.js, React/Next.js, Docker, Kubernetes, and advanced AI models like GPT-4, emphasizing scalability and real-time capabilities.

HuixiangDou is an open-source professional knowledge assistant powered by Large Language Models (LLMs), specifically designed to enhance group chat interactions with technical assistance. It employs a sophisticated three-stage pipeline for preprocessing, rejection, and response generation, ensuring relevant and concise answers without overwhelming chat participants. The tool offers real-time streaming chat capabilities and does not require extensive training, supporting CPU-only configurations with minimal memory requirements (2GB, 10GB). HuixiangDou provides a comprehensive suite of web, Android, and pipeline source code, making it industrially viable and commercially adaptable. It also features integrations with platforms like WeChat and Feishu, and supports various LLMs, retrieval methods, and preprocessing techniques, including hybrid knowledge graph and dense retrieval.

linkedin-mcp-server provides an open-source solution for connecting AI assistants, such as Claude, to LinkedIn. This tool allows AI agents to access various LinkedIn functionalities, including retrieving profile information with specific section selections (experience, education, skills, etc.), managing connection requests, listing and searching messaging inbox conversations, and sending messages. It also supports extracting company profiles and posts, searching for jobs with filters, and getting detailed job posting information. The server operates by controlling a real browser session, ensuring compliance with LinkedIn's terms of service by not exploiting undocumented APIs. Users can install it via uvx, as a Claude Desktop MCP Bundle, or through Docker, with options for both stdio and streamable HTTP transport modes.

just-prompt is a lightweight Model Control Protocol (MCP) server designed to offer a unified interface to various leading Large Language Model (LLM) providers, including OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama. It simplifies interaction with these diverse models by providing a consistent API for sending prompts, either as direct text or from files. Key features include the ability to run multiple models in parallel, automatic model name correction, and the option to save responses to files. The tool also supports advanced reasoning controls for specific models, such as OpenAI o-series, Claude, and Gemini, allowing users to fine-tune the depth of internal reasoning or thinking tokens/budget. This makes it ideal for developers and data scientists looking to experiment with or integrate multiple LLMs efficiently.

Palona AI offers a suite of restaurant-specific AI systems designed to streamline operations and enhance customer experience. Its AI host answers every call, takes complex orders, upsells naturally, and manages reservations, sounding like a trained team member. The platform also includes Operations AI for quality monitoring via security cameras and Catering AI to grow first-party catering. Palona integrates with existing systems like Toast, Square, and Yelp, allowing restaurants to go live in days with zero staff training. It uses generative AI for personalized, on-brand interactions, voice cloning for natural conversations, and never misses a call, boosting sales and guest satisfaction.

Artificial Studio is a comprehensive platform offering over 50 AI tools designed for creators and agencies to streamline their content creation workflows. It consolidates various AI capabilities into one place, enabling users to generate and edit images, videos, music, mockups, animations, avatars, and 3D models with ease. The platform aims to save time and money by eliminating the need for multiple subscriptions and tools, providing an intuitive workflow with a zero learning curve. Key features include text-to-video generation, image editing (background removal, scene extension), original music composition, 3D object conversion from 2D images, and creative extras like emoji design and face swapping. Artificial Studio also offers API access for businesses looking to integrate AI content generation into their own products or services.

ALIAgents.ai is a decentralized application (dApp) built on the AI Protocol, designed to revolutionize AI with blockchain technology. It enables users to explore and interact with expressive AI agents and swarms. A key feature is the world's first AI News Swarm, which delivers real-time, AI-driven journalism, aiming to provide instant and verified news without human bias or delays. The platform supports the creation, customization, and monetization of agentic AI, allowing for the development of custom generative and agentic AI systems. It leverages on-chain AI agents and generative AI to create a decentralized ecosystem for artificial intelligence.

Bahama provides an agent-first AI cloud infrastructure, enabling AI agents to deploy web applications, provision databases, and manage storage effortlessly. It integrates with existing AI coding tools like Claude Code, Cursor, and Codex via an MCP server, allowing agents to build and ship infrastructure-ready apps from the start. Bahama handles the orchestration behind the scenes, ensuring projects are isolated, resources are managed, and deployed apps run at scale. Key features include secure project management, managed SQL databases, storage buckets for assets, and secure secret management, all designed to automate the last mile of app development for AI agents.

Tianji is an open-source project dedicated to creating large language models adept at understanding and navigating traditional social scenarios. It provides comprehensive tutorials and implementations across various AI development aspects, including prompt engineering, RAG (Retrieval-Augmented Generation) knowledge bases, intelligent agent development, and LLM fine-tuning. The platform supports integration with popular models like ChatGPT, ZhipuAI, ERNIE, DeepSeek, InternLM, and Qwen, allowing for both online and local deployment. Tianji offers practical applications such as knowledge base dialogue systems for social etiquette, agent applications with tool-calling capabilities, and tools for data manufacturing and cleaning of social interaction corpora. It aims to equip users with full-stack application development skills for large language models.

WeClone offers a comprehensive, open-source solution for creating digital avatars from personal chat history. It allows users to fine-tune LLMs using their chat logs, including support for image modal data, to accurately replicate their unique communication style. The tool supports data export, preprocessing, model training, and deployment, with current integration for Telegram and planned support for WhatsApp, Discord, and Slack. WeClone emphasizes privacy through information filtering and localized fine-tuning, ensuring secure and controllable data. It also provides deployment options to various chatbot platforms like AstrBot and LangBot, enabling users to bring their digital self to life in conversational AI applications.

Envision provides AI-powered assistive technology for people who are blind or have low vision, enabling them to access visual information independently. The suite includes the Ally App, Ally Glasses, Envision Glasses, and the Envision App, all designed to articulate visual information into speech. Key features encompass real-time text recognition, scene description, document scanning, web search, and voice control. The Envision Glasses offer a hands-free experience for navigating the visual world, while the Ally App acts as an accessibility-first AI assistant for daily independence. The platform supports over 60 languages for text recognition and even handles handwriting, fostering knowledge, independence, and possibility for its users.