AI-103: Developing AI Apps and Agents on Azure

By Microsoft · Question Mode

QuestionQ150

VerifiedImplement generative AI and agentic solutions

You have a configurable guardrail in a Microsoft Foundry project.

A Python application has a text blocklist named ConfidentialTerms that already contains terms to detect in user messages. Each incoming user message is stored in input_text.

You plan to moderate input_text before sending the text to a language model.

You need to analyze input_text by using the existing blocklist.

How should you complete the code?

Select
...

blocklist_name = “ConfidentialTerms”

client = ContentSafetyClient(endpoint, credential)

request =  (

    text=input_text,

    blocklist_names=[blocklist_name],

    halt_on_blocklist_hit=False

)

analysis_result =  (request)

...
Explanation

AnalyzeTextOptions creates a text-moderation request and supports blocklist_names to supply the existing blocklist. client.analyze_text(request) submits that request for text analysis; image analysis and blocklist-management methods do not perform the required text moderation.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!