Compozy

What is Compozy?

Compozy is an advanced agentic orchestration platform designed to simplify the creation, deployment, and management of multi-agent AI systems. It uses declarative YAML to deliver scalable, reliable, and cost-efficient distributed workflows. Compozy is built on Go's high performance and Temporal's fault-tolerant reliability, making it a robust solution for enterprises looking to automate complex processes.

Benefits

Compozy offers several key advantages:

• Enterprise-Grade Reliability: Built on Temporal, a workflow engine trusted by Fortune 500 companies, Compozy ensures durable, fault-tolerant workflows that can withstand failures.
• Scalability and Security: Designed with Go at its core, Compozy enables secure, scalable AI workflows, giving developers greater control than no-code platforms.
• Open Source Advantage: Fully customizable and free from vendor lock-in, Compozy delivers robust enterprise features right out of the box.
• Full-Featured Orchestration: Supports AI agents, a broad spectrum of task types, and custom JS/TS tooling, seamlessly integrating with external systems.
• Language-Agnostic by Design: Uses a declarative YAML-based approach, making workflows easy for LLMs to learn and generate, with future multi-language support.

Use Cases

Compozy is versatile and can be used in various scenarios:

• Workflow Management: Design complex AI workflows using intuitive YAML templates with dynamic variables, directives, and Sprig functions for flexible orchestration.
• Agent Definition: Create intelligent agents with LLM integration, tool support, memory management, and structured outputs for sophisticated AI behaviors.
• Task System: Execute diverse tasks including basic operations, aggregates, collections, routers, signals, and waits with built-in error handling and parallel processing.
• Agnostic Runtime for Tools: Execute custom JavaScript/TypeScript code within tasks and agents using secure Bun runtime with granular permissions.
• MCP Integration: Discover and execute external tools via Model Context Protocol (MCP) servers through a secure proxy server.
• Signal-Based Events: Implement event-driven architectures with signal tasks, wait tasks, and triggers for decoupled workflow coordination and real-time responses.
• Scheduled Workflows: Create cron-job like automated workflows with built-in scheduling for recurring AI tasks and batch processing.
• Memory Management: Store and retrieve conversation history with configurable backends, flush strategies, and privacy controls for context-aware AI applications.

Pricing

Compozy offers flexible pricing options to suit different needs:

• Open Source: Free, full-featured workflow orchestration for individuals and teams building AI applications. Includes complete workflow engine, all task types, AI agent management, custom JS/TS tools, MCP server integration, self-hosted deployment, and community support.
• Cloud: Coming soon, managed cloud platform with enhanced features for growing teams and production workloads. Includes everything in Open Source plus managed infrastructure, automatic scaling, enhanced monitoring & observability, priority support, 99.9% uptime SLA, and team collaboration features.
• Enterprise: Custom pricing for advanced features, dedicated support, and custom solutions for mission-critical workloads. Includes everything in Cloud plus on-premise deployment options, advanced security & compliance, custom integrations, dedicated support team, SLA guarantees, and training & onboarding.