AI Agents & Automation

Planetoid is an open-source implementation of a graph-based semi-supervised learning method, as detailed in the ICML 2016 paper "Revisiting Semi-Supervised Learning with Graph Embeddings." It provides both transductive and inductive models for learning with graph embeddings. The tool is designed for machine learning researchers and data scientists who work with graph data, enabling them to perform tasks such as node classification and link prediction. It includes preprocessed datasets for Citeseer, Cora, and Pubmed, and offers clear examples for running both transductive and inductive versions of the model. The codebase is primarily in Python, making it accessible for those familiar with the language.

Awesome-Agent-Papers is a curated repository offering an extensive collection of research papers focused on Large Language Model (LLM) agents. It systematically organizes these papers into key categories such as agent construction, collaboration mechanisms, evolution, tools, security, benchmarks, and applications. This structured framework helps users understand the rapidly evolving field of LLM agents, from their architectural foundations to practical implementations. The repository aims to bridge fragmented research threads by highlighting connections between various agent design principles and emergent behaviors, making it an invaluable resource for researchers and practitioners alike who seek to stay current with the latest advancements in LLM agent technology.

ControlLLM is a framework designed to augment large language models (LLMs) with multi-modal tool utilization capabilities. This allows LLMs to tackle complex real-world problems by leveraging various tools and searching on graphs. The framework aims to enhance the automation and content generation potential of LLMs, enabling them to perform tasks that require more than just text-based understanding. While the live website currently indicates a maintenance message due to network issues, the underlying technology focuses on expanding the functional reach of LLMs through advanced tool integration.

RAG_Techniques is a comprehensive repository dedicated to advanced techniques for Retrieval-Augmented Generation (RAG) systems. It offers a dynamic collection of tutorials, each presented with a detailed notebook, to enhance the accuracy, efficiency, and contextual richness of RAG systems. The repository covers a wide array of methods, from foundational RAG concepts and query enhancements like HyDE and HyPE, to context enrichment techniques such as contextual chunk headers and relevant segment extraction. It also delves into advanced retrieval strategies, iterative techniques, and architectural patterns like Graph RAG and Self-RAG. The project aims to be a community-driven knowledge hub, fostering collaboration and innovation in the RAG field for researchers and practitioners alike.

Reflexion is a powerful tool designed for language agents, leveraging verbal reinforcement learning to enhance their capabilities. Associated with the NeurIPS 2023 paper, this repository offers comprehensive code, demos, and log files for various applications. It supports reasoning tasks, particularly with the HotPotQA dataset, allowing users to experiment with different agent types like ReAct and CoT, and various reflexion strategies. Additionally, it facilitates decision-making experiments using AlfWorld and programming tasks. While rerunning experiments with GPT-4 might be resource-intensive, the tool provides extensive logs from the paper's runs, making it an invaluable resource for researchers and developers in the AI agent space.

reggaetonBeGone is an experimental Raspberry Pi device designed to detect reggaeton music using machine learning and subsequently disrupt nearby Bluetooth speakers. The project focuses on edge ML, audio classification, and RF experimentation. It captures audio input, classifies the genre with an ML model trained on Edge Impulse, and triggers a Bluetooth interference routine when reggaeton is identified. The tool has evolved through versions, with v3.0 including on-device scanning, a strike system to reduce false positives, and an improved ML model. It requires hardware components like a Raspberry Pi, OLED display, push button, and Bluetooth audio receiver. The project is open source and provides instructions for building the device.

BoxMOT is an open-source tool designed for multi-object tracking workflows, offering a unified command-line interface (CLI) and Python API. It provides pluggable, state-of-the-art tracking modules with support for both axis-aligned and oriented bounding boxes. The tool streamlines direct tracking, cached benchmark evaluation, tuning, research loops, and ReID export, eliminating the need to rebuild detector and tracker stacks for each experiment. It supports Python versions 3.9 through 3.12 and includes swappable trackers with shared detector and ReID plumbing. BoxMOT is particularly useful for benchmark-oriented workflows, allowing for reusable detections and embeddings, and offers a public Python API for integration into applications and notebooks.

TaxyAI browser-extension is an open-source tool designed to automate browser interactions using advanced AI models like GPT-4. It enables users to provide ad-hoc instructions, allowing the AI to control the browser and perform repetitive actions. While currently in a research preview phase, it aims to support saved and scheduled workflows in the future. The extension operates locally, ensuring privacy by not sending page contents or instructions to external servers. It simplifies the DOM, identifies interactive elements, and uses an LLM to determine actions like clicking or setting values, making it a powerful tool for browser automation and task execution.

Ralph-loop-agent is an experimental package designed to bring continuous autonomy to the AI SDK. It implements the "Ralph Wiggum Technique," where an AI agent repeatedly attempts a task, receives feedback, and iterates until successful completion. Unlike traditional agentic workflows that stop after initial tool calls, Ralph-loop-agent wraps the AI SDK's `generateText` in an outer loop, allowing for verification, persistence, and feedback-guided retries. Key features include iterative completion, full AI SDK compatibility, flexible stop conditions (iterations, tokens, cost), built-in context management for long-running loops, streaming support, and feedback injection to guide subsequent attempts. This makes it ideal for complex, long-running tasks like code migrations or multi-file changes.

BMTools is an open-source repository designed for tool learning in large language models, providing a platform for the community to build and share tools. Users can easily create plugins by writing Python functions and integrate external ChatGPT-Plugins. The project is inspired by LangChain and optimized for open-sourced tools, aiming to be an academic version of ChatGPT-Plugins. It supports using single or multiple tools, developing customized tools locally, and contributing them to the BMTools repository. The platform also offers guidance on optimizing tool prompts for better AI model understanding.

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.

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.

ClaudeSync is an independent, open-source Python tool designed to automate the synchronization of local files with Claude.ai Projects. It offers seamless integration between your local development environment and Claude.ai, streamlining AI-powered workflows. Key features include cross-platform compatibility (Windows, macOS, Linux), extensive configuration options, and robust security measures to ensure data privacy. While it provides a one-way sync from local to Claude.ai, users should be aware that files not present locally will be removed from the Claude.ai project unless pruning is disabled. It supports Claude.ai Pro and Team plans, but not the Free plan, and requires Python 3.10+ and pip for installation. Users are advised to review Anthropic's Terms of Service before use, as ClaudeSync is not affiliated with Anthropic.

HandSpew is an innovative web application that allows users to generate thoughts and text through intuitive hand gestures. Leveraging MediaPipe for precise hand landmark detection, the tool translates your movements into words using the power of Gemini. This unique interaction method offers a fun and engaging way to explore AI-driven text generation. It's a free-to-use platform, making AI interaction accessible and enjoyable for anyone interested in novel human-computer interfaces. HandSpew provides a creative outlet for expressing ideas without typing, simply by moving your hands.

The Self-Hosting-Guide is a comprehensive resource designed to educate individuals and organizations on the intricacies of locally hosting and managing software applications. This guide delves into various aspects of self-hosting, including setting up on-premises and private web servers. Key topics covered include cloud solutions, Large Language Models (LLMs), WireGuard for secure networking, automation techniques, Home Assistant for smart home management, and general networking principles. It serves as an invaluable resource for anyone looking to gain control over their software infrastructure, offering insights into managing applications independently.

Self-Driving Pi Car is an open-source project designed for building a deep neural network-based self-driving car. It integrates Lego Mindstorms NXT for the physical robot structure and a Raspberry Pi 3 for computational power, allowing for real-time control and decision-making. The project provides comprehensive instructions for data collection, model training using TensorFlow, and deployment on the Raspberry Pi. Users can experiment with different architectures, learning rates, and optimizers to improve accuracy, and simulate the model before live deployment. It supports both Python 2 and Python 3 branches for flexibility.

The Synthetic Data Vault (SDV) is a Python library designed for generating tabular synthetic data. It employs various machine learning algorithms, from classical statistical methods like GaussianCopula to deep learning methods such as CTGAN, to learn patterns from real data and replicate them in synthetic datasets. SDV supports generating data for single tables, multiple connected tables, or sequential tables. Users can evaluate and visualize the quality of synthetic data against real data, diagnose problems, and generate quality reports. The library also offers features for preprocessing, anonymizing, and defining logical constraints to control data processing and improve synthetic data quality. SDV is part of The Synthetic Data Vault Project by DataCebo, providing a comprehensive solution for synthetic data generation and evaluation.

Clawlet is an ultra-lightweight and efficient personal AI assistant designed for seamless deployment and operation. It functions as a single static binary, eliminating runtime dependencies and CGO, and includes bundled SQLite with sqlite-vec for robust data management. This architecture allows for easy deployment on virtually any machine, providing immediate access to its powerful memory search capabilities. Clawlet supports various LLM providers including OpenAI, OpenRouter, Anthropic, Gemini, and local models via Ollama or vLLM. Key features include configurable agent generation defaults, optional semantic memory search with local embedding support, and secure defaults for file access and gateway binding. It also integrates with popular chat applications like Telegram, WhatsApp, Discord, and Slack, enabling versatile interaction with the AI assistant.

sd-webui-bilingual-localization is an extension designed to provide bilingual localization for the Stable Diffusion web UI. This tool enhances accessibility by offering bilingual translation, eliminating the need for users to search for original button labels or interface elements. It ensures compatibility with existing language pack extensions, meaning users do not need to re-import language settings. A key feature is its support for dynamic translation of title hints, providing a more intuitive user experience. The extension also includes advanced localization options such as Scoped and RegExp patterns, allowing for more flexible and precise translation rules. Scoped localization ensures that text is translated only within specific UI elements, preventing global pollution, while RegExp patterns enable powerful, rule-based translations. Installation is straightforward, either via URL within the web UI or by manual cloning to the extensions directory.

ShowUI is an open-source, end-to-end, lightweight vision-language-action model specifically designed for GUI agents and general computer use. It enables automation and interaction with graphical user interfaces through a unified model. The platform supports various functionalities including vllm inference for faster processing, API calling via Gradio Client without needing a GPU, and comprehensive training capabilities for grounding and navigation tasks across multiple datasets. ShowUI also offers features like UI-guided token selection, data annotation using GPT-4o, and integration into OOTB for local execution. It is suitable for researchers and developers working on AI-driven automation and GUI control.

SIF (Smooth Inverse Frequency) is an open-source project from PrincetonML that provides a simple yet effective baseline for generating sentence embeddings. The method utilizes a Smooth Inverse Frequency weighting scheme, which is crucial for creating robust sentence representations. The code is written in Python and requires libraries such as numpy, scipy, pickle, sklearn, theano, and lasagne. It includes functionalities for generating SIF embeddings, performing textual similarity tasks, and supporting supervised tasks like sentiment analysis. SIF is particularly useful for NLP researchers and practitioners looking for a reliable and easy-to-implement method for sentence embedding in various text analysis applications.

SJC (Score Jacobian Chaining) is an open-source research project that enables the generation of 3D models from pretrained 2D diffusion models. Presented at CVPR 2023, this tool leverages the chain rule on learned gradients and back-propagates the score of a diffusion model through the Jacobian of a differentiable renderer, specifically a voxel radiance field. This innovative approach aggregates 2D scores from multiple camera viewpoints into a cohesive 3D score, effectively repurposing existing 2D models for 3D data generation. The project addresses the technical challenge of distribution mismatch with a novel estimation mechanism and supports off-the-shelf diffusion image generative models, including Stable Diffusion. It also includes implementations of the Karras sampler and a customized voxel NeRF.

Serenata de Amor is an open-source project that leverages artificial intelligence for the social control of public administration. The core of the project is Rosie, an AI that analyzes Brazilian congresspeople's expenses to identify suspicious spendings. To make this data accessible, Jarbas provides a web interface where users can visualize expenses and investigate suspicions. The project aims to empower citizens with data, focusing on smart citizens, accountability, and open knowledge. While the project is not frequently updated, it provides valuable tools for civic innovation and encourages community contributions to resolve bugs and propose improvements.

Mistral Nemo was an AI chatbot available on Hugging Face Spaces, designed to facilitate task automation and content generation. The tool aimed to provide an accessible platform for exploring AI chatbot capabilities, particularly for educational use and interactive experiences. However, the application is currently experiencing a runtime error, indicating that the underlying repository for the model is not found or accessible. This suggests that while the tool was intended to offer free exploration of AI chatbot functionalities, it is presently non-functional due to issues with its model's availability or authentication.