Compozy

Compozy is a next-level agentic orchestration platform designed to create, deploy, and manage robust multi-agent AI systems. It unifies agents, tasks, tools, and signals into scalable YAML workflows. Powered by Go and Temporal, it ensures high performance, reliability, and fault tolerance, while optimizing costs and providing enterprises with full infrastructure control. Compozy simplifies complex fan-outs, debugging, and monitoring for production-ready automation of distributed AI workflows.

## How to Use

Compozy is used by defining AI workflows, agents, and tasks through intuitive declarative YAML templates. Users can design complex AI workflows with dynamic variables and functions, create intelligent agents with LLM integration and tool support, and execute diverse task types including basic, parallel, and collection operations. It supports custom JavaScript/TypeScript tools via a secure Bun runtime, integrates with external systems using Model Context Protocol (MCP), handles signal-based events, and manages scheduled workflows and memory for context-aware AI applications.

## Key Features

- Declarative YAML Workflows - Multi-agent AI System Orchestration - Scalable and Reliable Execution (Go & Temporal) - Advanced Task System (Basic, Parallel, Collection, Router, Signal, Wait) - Custom Tooling with Agnostic Runtime (Bun for JS/TS) - Model Context Protocol (MCP) Integration - Signal-Based Event Handling - Scheduled Workflows - Memory Management for Agents

## Use Cases

- AI-powered Go code analysis workflows - Automated daily code analysis and repository scanning - Implementing event-driven architectures for decoupled workflow coordination - Batch processing of files or data - Creating context-aware AI applications with conversation history