LLM and agent features¶

These settings change how a model call or an agent run behaves, independent of which operator you use. Each is a parameter on the operator or decorator.

Structured output and XCom: output_type returns a typed Pydantic object through XCom instead of a string.
Multi-turn sessions and message history: message_history carries a conversation across agent runs.
Capabilities and guardrails: capabilities= adds pydantic-ai capabilities such as Thinking and WebSearch, and pydantic-ai-shields guardrails, to an agent.
Code mode: the CodeMode capability lets the model call several tools from one Python snippet instead of one round trip per call.
Approval gates for LLM operators: require_approval=True pauses an LLM operator until a person approves, edits or rejects the output.
Human-in-the-loop (HITL) review for agents: enable_hitl_review=True opens an iterative review loop on an agent, with a chat UI and REST API for the reviewer.
Approve an agent’s tool calls: a tool marked with pydantic-ai’s approval API pauses the agent before the call runs, until a person approves or rejects it.

The last three are different tools for different jobs: an approval gate is a one-shot decision on one output, a HITL review is a conversation with a running agent, and a tool approval is a decision on one action before it happens.

Making retries cheap with durable=True is a reliability feature and lives under Reliability and operations.