Open-source platform to build and deploy AI agent workflows.

A self-contained, lightweight workflow engine for enterprise and small teams. Single binary with Web UI. 100% open source. No vendor lock-in. It natively supports running containers and executing commands over SSH. Offline or air-gapped environment ready.

Blades is a Go-based multimodal AI Agent framework.

Browser Operator - The AI browser with built in Multi-Agent platform! Open source alternative to ChatGPT Atlas, Perplexity Comet, Dia and Microsoft CoPilot Edge Browser

(ACL-2025 main conference) SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing

ArxivFlow - Periodic Track on arXiv Paper

(ACL-2025 main conference) Dolphin: Moving Towards Closed-loop Auto-research through Thinking, Practice, and Feedback

Build AI-powered engineering tools in simple YAML

This repository contains showcase projects developed by ii-agent.

Robota is an AI agent framework written in JavaScript/TypeScript.

JVM-based Multi-Agent Orchestration Framework - The Spice

Markdown-based documentation for all modules and components in the Summoner codebase

An 🧪 experimental, 🧭Parlant's ARQ (Attentive Reasoning Queries) into 🔷Microsoft Agent Framework via middleware.

Project uses Gemini LLM and Vision-Language Models (VLM) to automatically generate presentation slides from documents. It streamlines content summarization and design, enabling efficient slide creation by interpreting both text and visual elements in the source material. (SuZeAI mentor)

Transform ideas into production ready code through the use of specialised AI agents working in coordinated phases.

IntentusNet — Language-Agnostic AI Agent Runtime & Orchestration Framework A high-performance, secure, multi-agent runtime that supports intent routing, fallback, workflow orchestration, and optional EMCL encryption. Designed for scalable AI systems, distributed automation, and MCP-compatible agent workflows.

Though this project is about reflection in the generation of tweets it is a foundational piece in having your LLM reflect on what it has created and producting something better. I see this for use in blogs, articles, tweets, emails, etc..

In this repo I have made Devil(Shetani) calculator assigned by Sir Zia, was great fun to make this Agent.

Advanced AI Assistant with conversational memory, agent capabilities, MCP , knowledge bases, multi-user support, and RAG (knowledge graph + vector search). Includes prompt management and other AI services.

See AIs debating in a mock trial