FAQs

Answers by
service.

Practical answers for every IaGenify service.

FAQAgent1 min read

Does the agent cost extra?

No. The agent, its tools, and its multimodal capabilities are included on every plan, including the free one. You simply pay for the underlying usage credits.

Read answer
FAQAgent1 min read

What can the agent access?

Only what you explicitly permit. It can read your managed databases and connected storages, but it will always ask for approval before executing a write or delete operation.

Read answer
FAQAgent1 min read

Can I change the model?

Yes, the agent can transparently route requests to different models depending on the complexity of the task and the required capabilities.

Read answer
FAQAgent1 min read

Can I require a structured response?

Yes. Define an output schema when you create an assistant session to describe the shape the response should follow.

Read answer
FAQAgent1 min read

Can an agent remember earlier turns?

A session created with memory enabled keeps conversation history for follow-up turns; an ephemeral session does not.

Read answer
FAQAgent1 min read

What makes a good task for an agent?

Give the agent a clear outcome, the relevant context, and boundaries for decisions that need your review.

Read answer
FAQAgent1 min read

How can I understand why an agent run stopped?

Review the run's step history and stop reason, then compare the last completed action with the result you requested.

Read answer
FAQAgent1 min read

Can I ask the agent to draft an email without sending it?

Yes. Ask for a draft and make clear that you want to review it before any message is sent.

Read answer
FAQAgent1 min read

Can I send an image to an agent along with text?

Yes, image inputs can accompany text when the selected model supports them; describe what the agent should inspect in the image.

Read answer
FAQAgent1 min read

Can I review an agent's proposed changes before they are applied?

Yes. Choose a permission mode that asks before a tool action, and review the proposed operation before allowing it.

Read answer
FAQAgent1 min read

What happens if an agent's selected model is unavailable?

If fallback models are configured, the agent can try one of those alternatives; review the final result because another model may respond differently.

Read answer
FAQAgent1 min read

Why did an agent stop before completing every requested step?

An agent run can stop when it reaches its configured step limit, encounters a blocking error, or is stopped; review the run details to identify which occurred.

Read answer