Why is "Verify the retriever's filter criteria and data source connection" the correct answer?

If agent answers are not satisfactory when using PDF files as a knowledge source, the issue is likely caused by:

Retriever misconfiguration

If filters are too broad or too restrictive, AI may fail to find relevant information.

Checking filter logic and retrieval scope helps improve accuracy.

Incorrect data source connection

If the retriever is not properly linked to the PDF storage location, it may fail to retrieve relevant information.

Ensuring a stable connection between Salesforce Data Cloud and the retriever prevents retrieval failures.

Parsing Issues with PDF Files

If PDFs are not properly indexed, AI may struggle to extract relevant content.

Ensuring structured document formatting improves AI comprehension.

Why Not the Other Options?

❌ A. Analyze the data mapping between source fields and Data Cloud object fields.

Incorrect because data mapping issues primarily affect structured CRM data, not PDF-based knowledge sources.

The issue likely stems from retrieval settings, not field mapping.

❌ B. Check that the agent has the PDF file field permission access for the data library.

Incorrect because permission access issues would prevent AI from accessing PDFs entirely rather than causing poor response quality.

AI can still generate responses, even if they are inaccurate, which means the issue lies in retriever settings, not permissions.

Agentforce Specialist References

Salesforce AI Specialist Material details how retriever filters and data sources impact AI-generated answers​.

Salesforce Certification Guide mentions the importance of verifying retriever configurations for accurate knowledge retrieval​.