Skip to main content

​
Description

Use LLM Data when you need to extract one value from the current conversation. It is best for a single name, email, number, date, appointment time, reference number, or category. Use LLM Custom instead when you need several fields from the same answer.

​
Manual Inputs

​
Type

Tells the tool what kind of answer to look for and how to format it.

​
Reasoning

How hard the model works to interpret the answer.
  • low — direct answers (“My email is jane@example.com”)
  • medium — the caller may imply the value or answer less directly
  • high — genuinely subtle answers needing careful interpretation (can be slower)

​
Label

The short name of the value being extracted. Keep it noun-like; put rules in Description.

​
Description

The main instruction. State exactly what counts, what does not, and how to format the result. This takes priority over the general type rules.
For dates and datetimes, be explicit about whether partial answers are allowed. For reference numbers, say what similar values should be ignored.

​
Options

Used only for struct-enum. Provide the allowed choices the tool can return. Use stable, machine-friendly values:
If you use labels or descriptions for the options, keep them short and distinct.

​
Destination

Optional variable to write the extracted value into. Leave blank if you only need the result boolean.

​
Manual Outputs

​
Conditions

​
Success (true)

A valid value was found and, if a destination was configured, written successfully.

​
Failure (false)

The value was missing, unclear, irrelevant, invalid for the selected type, or could not be written.

​
Variables this tool writes

LLM Data does not own a variable namespace. It writes exactly one variable — the Destination you map — and only when a valid value is extracted. The write is entirely yours to define: point Destination at a custom scratch variable such as {{custom.customer_sentiment}}, or at a global such as {{contact.email}}. Leave Destination blank and the tool writes nothing to any variable; it returns only the result boolean for routing. The data type of what is written follows the Type you choose:
  • struct-text → e.g. {{contact.email}} (string) = jane@example.com
  • struct-num → e.g. {{custom.claim_amount}} (number) = 50000
  • struct-date → e.g. {{custom.incident_date}} (string, YYYY-MM-DD) = 2026-07-14
  • struct-datetime → e.g. {{custom.appointment_datetime}} (string, ISO with the team timezone offset) = 2026-07-18T14:30:00+10:00
  • struct-enum → e.g. {{custom.customer_sentiment}} (string, the matched option value) = negative
The Destination must be a well-formed variable reference — a custom scratch variable (any {{custom.*}} path works without prior setup), or a declared global. See Variables for the full list and how references resolve.

​
Example Usage

Extract an email:
Extract an appointment time:
Extract sentiment:

​
Common Issues

  • Enum extraction fails — check that Options is filled in and the answer can reasonably match one option.
  • Date or datetime fails — check whether the caller actually gave enough information.
  • Returns false even though the value was extracted — check the Destination variable reference.
  • Inconsistent results — make Description more specific.

​
Next Steps