certslothcertsloth
PAA/Topic 01

Google Cloud / Professional · Upcoming

Low-code agents and connected knowledge

2 min read5 recall promptsReviewed 2026-10-10

Memory hook: Flow for certainty; generation for flexibility.

Must remember

  • Gemini Enterprise Agent Designer and Customer Experience Agent Studio provide low-code ways to configure agent behavior. Use explicit states/pages, transitions and event handling when the process requires predictable order.
  • Instructions define role, scope, output and escalation. Few-shot examples demonstrate behavior; prompt templates should distinguish trusted instructions from user or retrieved content.
  • Enterprise connectors and Agent Search retrieve organizational knowledge. Preserve document permissions, tenant boundaries, refresh schedules and source attribution throughout ingestion and retrieval.
  • Multimodal sources need appropriate extraction, metadata, chunking and evaluation. A transcript or image caption can lose important context; verify against representative source material.
  • A conversational workflow should collect missing information, validate it, invoke permitted tools and confirm outcomes. Escalate when evidence, authorization or task confidence is insufficient.
  • Test normal, ambiguous, adversarial and unavailable-data paths. A low-code designer reduces implementation work but does not remove security, reliability or evaluation responsibilities.

Choose under exam pressure

Requirement Choice and reason
A regulated process has mandatory ordered steps Explicit workflow states with controlled generative assistance.
An employee asks about private policies Authorized enterprise retrieval with evidence and access checks.

Traps

  • A connector does not make every indexed document public to every agent user.
  • A fluent conversation can still skip a required business step.

Active recall

1. When prefer deterministic transitions?

When order, required fields and allowed actions must be predictable.

2. Why test multimodal extraction?

OCR/transcription or interpretation errors can corrupt the retrieved evidence.

3. What should an agent do without adequate evidence?

Ask for clarification, abstain or escalate according to the workflow.

4. Why version prompt templates?

To reproduce behavior and compare changes against an evaluation baseline.

5. What must follow a tool call?

Validate the result and report the actual outcome rather than assume success.

Sources

CLOSE THE NOTES. EXPLAIN THE CHOICE.

How well could you recall it?

Your next review is based on this answer. Progress stays in this browser.

Search across every published topic.