Agents Overview

Agents are autonomous AI workflows that execute multi-step logic, call LLMs, use tools, and compose with other agents.

Agent Types

Choose the type that best fits your use case.

Conversational interfaces for customer support, Q&A assistants, and interactive experiences. Handle multi-turn conversations with context.

Transform long content into concise summaries. Perfect for document digests, meeting notes, and content processing.

Analyze data to provide personalized suggestions. Use for product recommendations, content curation, and decision support.

Process and interpret data to generate insights. Build report generators, trend analyzers, and data processors.

For use cases that don't fit standard categories. Full flexibility to design any workflow.

Every agent is built from these core elements.

The visual flow of nodes that defines your agent's logic:

Entry Point - Where execution starts, defines inputs
Agent Steps - Call AI models with context, with optional integrations as tools
Exit Point - Define what your agent returns

Inputs your agent accepts. Define types, validation rules, and defaults. Make agents reusable across different contexts.

External capabilities your agent can use: