Flash Cards · Implementation

Flash Card: Letting an LLM Take Actions

July 31, 2026 · 3 min read

Exam-style

A customer assistant on Bedrock currently only answers questions. It now needs to actually reschedule a delivery by calling an internal REST API, then tell the customer what it did. What wires that up?

Reveal the answer

E. Define a Bedrock Agent with an action group backed by a Lambda function, so the agent invokes it and feeds the result back into the conversation

Action groups are the supported wiring for side effects: the model plans, the agent calls your Lambda or returns control to your application, and the result comes back into the conversation for the next turn. Prompting the model with an API spec and parsing its reply rebuilds that orchestration by hand, without the schema validation or the return path. Retrieving API documentation gives the model knowledge about the API rather than the ability to call it, and guardrails filter content instead of performing actions.

Generative AI Development · part of The Exam Room

Q. Your LLM needs to take real actions, like calling an API. What wires that on Bedrock?

A. Bedrock Agents with action groups: the model plans, the agent invokes your Lambda (or returns control to your app) to perform the action, and the result feeds back into the conversation.

Why? Side-effecting actions belong behind action groups, not baked into the prompt.

These posts are LLM-aided. Backbone, original writing, and structure by Craig. Research and editing by Craig + LLM. Proof-reading by Craig.