Prompt Engineering20% of the exam

Prompt template types: field generation and flex

What each prompt template type is bound to, with the considerations the exam tests for field generation (one object, one writable text field, a reviewed Einstein draft on a Dynamic Forms record page) and flex (up to five inputs you define, any caller), and how to pick the type from a requirement.

Official exam objective
Identify the considerations for using a prompt template type such as field generation and flex types.

~8 min read · 8 practice questions ·

What the exam expects

  • Pick a prompt template type (sales email, field generation, record summary or flex) from the requirement's output, inputs and entry point.
  • State the field generation considerations: one object and one writable text field chosen at creation, an Einstein button on a Dynamic Forms record page set up in Lightning App Builder, and a draft the user reviews before saving.
  • State the flex considerations: no predefined entry point, up to five inputs you define (objects or free text), and invocation from Flow, Apex, the REST API or an agent.
  • Recognise when two requirements need two templates because the type is fixed and bound to its entry point.
  • Key terms: template type (the kind of template chosen at creation, which fixes its inputs and entry point); field generation (fills one record field from a button on the record page); flex (general-purpose type with inputs you define); input/resource (a record or free-text value the template receives); Dynamic Forms (record pages built from individual field components).

Key facts

Field generation objects
Detail
Supports only one object; use flex for several
Field generation targets
Detail
Writable text, text area, text area (long), text area (rich); not read-only fields
Field generation placement
Detail
Lightning App Builder, Dynamic Forms field, Einstein Generative AI section of field properties
Field generation at run time
Detail
User reviews the draft, clicks Use, edits and saves
Knowledge rich text areas
Detail
Field generation unsupported on __kav rich text areas; use flex, Flow or Apex
Flex inputs
Detail
Maximum 5
Flex input setup
Detail
Each input has a unique Name and API Name; can be marked required
Free-text input
Detail
Flex only, up to 127,000 characters
Flex callers
Detail
Apex, Flow, REST API or Agentforce; no predefined entry point
Flex in a flow
Detail
A flow can fill a field with it automatically, with no human review
Record summary
Detail
Used by the Summarize Record standard invocable action and agent action
Sources for this section (10)
  • Why does a requirement with several separately defined object inputs call for flex rather than field generation?
  • When does generated text reach a field without anyone reviewing it?
  • How many inputs can a flex template define?
  • Which field types can a field generation template populate?

The questions are free to read. Answers with their sources and self-grading are included with the pack.

Keep reading with the Agentforce Specialist pack

  • Core concepts
  • Exam traps
  • Worked scenario

The pack includes:

  • All 26 lessons of the Agentforce Specialist guide, with reading progress
  • Interactive decision walks and step-by-step sequence steppers
  • Inline practice questions and one-click topic practice (8 questions for this lesson)
  • An official source cited for every claim
  • An AI tutor that answers from the guide
  • Unlimited timed practice exams

Practice for the Agentforce Specialist exam