Agent steps

An agent step applies an instruction to every record of its input and returns the fields declared in output.

- id: classify
  agent: Classify the expense category from the description.
  input: invoices
  output:
    number: str
    category: str?
    tags: list[str]?
  llm: default
  concurrency: 4

Output schema

Types: str, int, float, bool, list, dict, list[str] (or any of the scalar types). A trailing ? makes the field optional.

The model's answer is validated against the schema. A record that fails validation is retried once, then rejected: the run continues and an item.rejected event reports how many records failed.

Providers

llm:
  default: { provider: anthropic, model: claude-opus-5-5 }
  offline: { provider: echo }

Provider	Notes
anthropic	Claude with structured output. Needs agentl[anthropic] and an API key.
echo	Offline. Copies matching fields from the record. Useful for tests and dry runs.

Each agent step chooses a provider with llm:. Register your own with register_provider.

Data privacy

Agent steps send record contents to the configured provider. Use a local provider for data that must not leave your machine, and select only the fields the agent needs with a select step first.