joanmarcriera/Home-office-automations
Agent
Goose is an open-source, extensible AI agent designed to go beyond simple code suggestions. It is built to install, execute, edit, and test code autonomously or with human supervision, using any LLM that supports tool-calling. Hosted by the Agentic AI Foundation (AAIF), it serves as a robust platform for building and…
joanmarcriera/Home-office-automations
Agent
GPT Researcher (v4.5+, early January 2027) is an autonomous agent designed for comprehensive online research on any given topic. It plans the research, browses the web, and synthesizes a final report with deep citations. It uses a "master-agent" and "research-agent" pattern to break down complex queries into…
joanmarcriera/Home-office-automations
Agent
Home Admin Agent Tools are the local service adapters exposed to the Ralph home-admin agent for interacting with household infrastructure. They wrap the complex REST APIs of household services into simplified, agent-discoverable tools following the Model Context Protocol (MCP) / FastMCP 3.1 specification. These tools…
joanmarcriera/Home-office-automations
Agent
Directory of tools, frameworks, and runtime environments for building, running, and orchestrating autonomous AI agents.
joanmarcriera/Home-office-automations
Agent
Letta (v1.15.x+, early January 2027) is a framework for creating stateful AI agents with "infinite" memory. It manages memory as a tiered system (long-term, short-term) to overcome LLM context window limits by treating the context window as a "cache" for a larger, persistent memory store, now natively supporting the…
joanmarcriera/Home-office-automations
Agent
Mem0 (v2.5+, early January 2027) is a specialized memory and personalization layer for AI agents and applications that stores, prioritizes, and retrieves durable user, task, and workflow context over time. Often described as the "long-term memory" of the agentic stack, it enables agents powered by Claude 5.6, GPT-5.6…
joanmarcriera/Home-office-automations
Agent
Multi-Agent Systems (MAS) represent an architectural pattern and execution paradigm where multiple autonomous AI agents—each possessing distinct roles, tools, and context windows—collaborate, negotiate, and coordinate to solve complex, multi-step problems. As of early 2027, Multi-Agent Systems form the foundation of…
joanmarcriera/Home-office-automations
Agent
MultiOn is an enterprise-grade AI agent framework, SDK, and API designed specifically for autonomous web navigation, interaction, and stateful browser control. As of late December 2026, it operates on the mature API v3 architecture, serving as a high-performance "motor cortex" for AI. It enables frontier models like…
joanmarcriera/Home-office-automations
Agent
NVIDIA NeMo Retriever is a family of generative AI microservices (v2027.1.x+, early January 2027) designed to provide high-performance, agent-ready retrieval-augmented generation (RAG) capabilities. It enables organizations to connect their custom models to live enterprise data and deliver highly accurate…
joanmarcriera/Home-office-automations
Agent
NVIDIA NeMo Claw is an enterprise-grade agent orchestration framework designed for building, deploying, and managing high-performance AI agents. As of early January 2027, it serves as the primary agentic layer within the NVIDIA NIM ecosystem, optimized for the NVIDIA Rubin, Ultra-Blackwell, and Grace Hopper…
joanmarcriera/Home-office-automations
Agent
Open Agents is an open-source framework and collection of deployable AI agents created by Vercel Labs. It is designed to enable "Computer Use" and "Web Use" capabilities, allowing autonomous agents to navigate the web, interact with authenticated SaaS applications, run code in sandboxes, and perform complex multi-step…
joanmarcriera/Home-office-automations
Agent
The Perplexity Agent API is a suite of programmatic interfaces providing developers with access to Perplexity's agentic workflows and orchestration capabilities. As of early January 2027, it features specialized models like Sonar Pro, Sonar Reasoning Pro, and Sonar Deep Research, which integrate real-time web search…
joanmarcriera/Home-office-automations
Agent
Phidata is a Python-native framework for building AI assistants with memory, knowledge, and tools. As of early January 2027, Phidata has fully transitioned and rebranded into the Agno framework (v3.x). It serves as a primary enterprise bridge for transforming raw LLMs into stateful, autonomous agents, offering native…
joanmarcriera/Home-office-automations
Agent
Replit Agent (v5, late November/December 2026) is an autonomous, natural-language software engineering agent fully integrated within the cloud-based Replit development workspace. Unlike generalized coding assistants, Replit Agent operates as a high-autonomy developer that can provision full-stack workspaces, configure…
joanmarcriera/Home-office-automations
Agent
Roo Code is an open-source, AI-powered autonomous coding agent for VS Code, JetBrains IDEs, and CLI workflows. Originally forked from Cline, it has evolved into a customizable platform that supports specialized "Custom Modes," deep Model Context Protocol (MCP) integration, and multi-model orchestration. In early 2027…
joanmarcriera/Home-office-automations
Agent
Superpowers is a comprehensive software development workflow and agentic skills framework designed for next-generation coding agents like Claude Code, Cursor, and Aider. It builds on top of composable "skills" to enforce a rigorous engineering process, optimized for frontier models like Claude 5.1 and GPT-5.5 while…
joanmarcriera/Home-office-automations
Agent
Symphony is an enterprise-grade autonomous implementation framework open-sourced by OpenAI (updated early January 2027) designed to transform structured project requirements into fully isolated, self-verifying autonomous implementation runs. It automates high-level work items (such as Jira or Linear issues) by…
joanmarcriera/Home-office-automations
Agent
WorldClaw is an open-source agentic world-model framework and environment simulation toolkit developed by Tencent. Released in August 2026, WorldClaw combines spatial-temporal physical world modeling with multi-agent orchestration, enabling autonomous agents to simulate, predict, and execute actions across complex…