Introduction to Tensor

Tensor is a powerful AI Agent Automation Platform that enables you to create, manage, and orchestrate AI agents with ease. Built with modern technologies and designed for scalability, Tensor provides a robust foundation for building AI-powered applications.

Key Features

• Isolated Agent Runtimes: Each agent operates in its own isolated environment, ensuring security and reliability
• Flexible Skill System: Define and implement agent capabilities using JSON schemas
• Advanced Workflow Engine: Create complex automation workflows with typed definitions
• Knowledge Management: Built-in vector database for efficient knowledge storage and retrieval
• Modern Stack: Built with Next.js 14+, TypeScript, and other cutting-edge technologies

Platform Components

Tensor is organized into several core components:

• Chat: The primary interface for interacting with agents
• Tools: Components that extend your agents' capabilities
• Memory: System for retaining and recalling information
• Data: Database system for storing agent-specific information
• Triggers: Automation system for event-based actions
• Keys: Secure management of API keys and credentials

Architecture Overview

Tensor is built with a modular architecture that emphasizes:

• Type safety and compile-time checks
• Secure inter-agent communication
• Scalable data storage with PostgreSQL and pgvector
• Real-time monitoring and observability
• Extensible skill system

Getting Started

To get started with Tensor, check out the following guides:

1. Installation Guide
2. Quick Start Tutorial
3. Core Concepts

Next Steps

• Follow the Installation Guide to set up your development environment
• Learn about Agents and how they work
• Explore the Skills System to extend agent capabilities
• Understand Workflows for complex automation