koala73/worldmonitor
World Monitor is a real-time global intelligence dashboard for AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface.
Awesome AI Agents › LLM-Powered Platforms
inngest/agent-kit
AgentKit is a TypeScript framework designed to build multi-agent networks with deterministic routing and rich tooling through the use of MCP (Multi-Channel Protocol). It enables developers to create complex AI agent systems that collaborate and share state in a controlled and fault-tolerant manner. The core concepts of AgentKit include Agents, Networks, State, Routers, and Tracing. Agents represent LLM (Large Language Model) calls combined with prompts, tools, and MCP. Networks allow multiple agents to collaborate with a shared state and support handoff between agents. The State is a fully typed state machine combined with conversation history, which is used for routing decisions. Routers provide autonomy in the system, ranging from code-based to LLM-based orchestration, allowing flexible and deterministic routing of tasks among agents. Tracing tools help debug and optimize workflows both locally and in the cloud. AgentKit supports multiple model providers and integrates with the Inngest Dev Server for local development and cloud deployment, ensuring fault tolerance. The framework encourages starting with code-based routing for complete control over network execution flow, enabling deterministic routing based on state updates. It also supports advanced patterns like MCP for rich tooling and multi-agent collaboration. The project provides examples and documentation to help users get started quickly, including an MCP agent example using the Neon database MCP server. Overall, AgentKit is a powerful tool for building sophisticated multi-agent AI systems with deterministic routing, shared state management, and extensive tooling support, aimed at TypeScript AI developers and teams.
https://github.com/inngest/agent-kit
Also in LLM-Powered Platforms
hiyouga/LlamaFactory
LLaMA Factory is a unified platform for efficient fine-tuning and deployment of over 100 large language and vision-language models, supporting advanced training methods, scalable resource management, and fast inference with broad industry adoption.
FoundationAgents/OpenManus
OpenManus is an open-source platform that enables users to create and run intelligent agents using large language models without restrictive access, supporting multiple agents and easy configuration for diverse AI applications.
hugohe3/ppt-master
AI-powered tool that generates natively editable PowerPoint presentations, complete with animations and audio narration, from any document, supporting custom .pptx templates.
zhayujie/CowAgent
CowAgent is an open-source, multi-model AI assistant and agent harness that autonomously plans tasks, utilizes skills, and learns with memory and knowledge, supporting various LLMs and communicatio...
janhq/jan
Jan is an open-source, privacy-focused ChatGPT alternative that runs large language models locally on your computer with optional cloud integration and custom AI assistant capabilities.
MadsLorentzen/ai-job-search
AI Job Search is a powerful, local AI application framework leveraging Claude Code to automate job application tasks, including evaluating postings, tailoring CVs and cover letters, and preparing f...
agno-agi/agno
Agno is a lightweight, model-agnostic library for building advanced AI agents with memory, knowledge, tools, and reasoning capabilities, supporting multi-modal inputs and outputs and featuring an advanced multi-agent architecture.