SWE-agent is an open-source tool that utilizes large language models (LLMs) like GPT-4o and Claude Sonnet 3.5 to autonomously fix bugs in GitHub repositories, solve cybersecurity challenges, and perform complex tasks. It features a mode called EnIGMA for offensive cybersecurity and prioritizes simplicity and adaptability.
Notte is an open-source browser using an agent, designed to improve speed, cost, and reliability in web agent tasks through a perception layer that structures webpages for LLM consumption. It offers a full stack framework with customizable browser infrastructure, web scripting, and scraping endpoints.
This article details six practical use cases for Model Context Protocol (MCP) to automate workflows using AI agents and integrations with tools like Slack, Google Calendar, BigQuery, Linear, GitHub, and HubSpot. It highlights the impact of these automations on team efficiency and productivity.
This article explores the Model Context Protocol (MCP), an open protocol designed to standardize AI interaction with tools and data, addressing the fragmentation in AI agent ecosystems. It details current use cases, future possibilities, and challenges in adopting MCP.
Browser Use is a library that enables AI agents to interact with web browsers, making websites accessible for automated tasks. It includes features for browser automation, agent memory, and various demos showcasing its capabilities.
The article discusses the emergence of AI agents in enterprise IT, highlighting Orby's development of Large Action Models (LAMs) designed for automating complex workflows. These models, unlike traditional LLMs, process actions such as application interactions and automate tasks in enterprise environments like Salesforce and SAP. The concept of 'traces,' sequences of actions for specific tasks, is used to fine-tune LAMs, and Orby's AI agent software stack allows for customization and scaling by technical personnel.
Zero Human Code: What I Learned from Forcing AI to Build (and Fix) Its Own Code for 27 Straight DaysAn experiment in agentic AI development, where AI tools were tasked with building and maintaining a full-service product, ObjectiveScope, without direct human code modifications. The process highlighted the challenges and constraints of AI-driven development, such as deteriorating context management, technical limitations, and the need for precise prompt engineering.
A summary of personal experiences using generative models while programming, highlighting the benefits and practical applications of LLMs in productivity and programming tasks.
GitHub Models now allows developers to retrieve structured JSON responses from models directly in the UI, improving integration with applications and workflows. Supported models include OpenAI (except for o1-mini and o1-preview) and Mistral models.
This article explores automating the process of converting scientific code into LaTeX documents using GPT models and Python, aiming to streamline documentation workflows in scientific projects.
