Master how Bedrock Agents orchestrate tools and knowledge.
0 / 5 completed
1 / 5
At standup, the team builds an orchestrating AI on Bedrock that plans and calls tools. What is this entity?
A Bedrock agent orchestrates a foundation model to break down user requests, call tools, and query data. It manages the reasoning loop for you. It is the top-level construct for agentic apps on Bedrock.
2 / 5
During a design review, the team exposes business APIs the agent can call. Which Bedrock construct defines these?
An action group defines the operations an agent can invoke, mapping API schemas to a backing Lambda or function. The agent picks actions based on the user's intent. This is how agents take real-world actions.
3 / 5
In a PR review, the team wants the agent to ground answers in company docs. Which Bedrock feature?
A Bedrock knowledge base provides retrieval-augmented grounding by indexing your documents in a vector store. The agent retrieves relevant chunks to answer with citations. This reduces hallucination on domain questions.
4 / 5
During a code review, a dev passes per-conversation context not visible to the user. Which mechanism fits?
Session attributes carry contextual data across turns of a conversation, available to the agent and your Lambda. They hold state like a user id without showing it in the prompt. This keeps conversational context consistent.
5 / 5
An incident report shows an action group failing to execute. Which AWS service typically backs the action?
An action group is commonly backed by a Lambda function that implements the declared operations. When the agent selects an action, Bedrock invokes the Lambda with the parameters. A failing action usually means a bug or permissions issue in that Lambda.