CCARF Practice Question

Paper 5 · Question 12

D5 Context Management & Reliability · Paper 5

[Scenario: Multi-agent Research System] In testing, the combined output of the web-search agent (85K tokens including page content) and the document analysis agent (70K tokens including chains of thought) totals 155K tokens, but the synthesis agent performs best with inputs under 50K tokens. Which solution is most effective?

• A.Modify upstream agents to return structured data (key facts, quotes, relevance scores) instead of verbose content and reasoning.
• B.Add an intermediate summarization agent that condenses findings before passing them to synthesis.
• C.Have the synthesis agent process findings in sequential batches, maintaining state between calls.
• D.Store findings in a vector database and give the synthesis agent search tools to query during its work.

Show answer & explanations

Correct answer: A

A: Modifying upstream agents to return structured data fixes the root cause by reducing token volume at the source while preserving essential information. It avoids passing bulky page content and reasoning traces that inflate tokens without improving the synthesis step.

B: An extra summariser can drop the quotes and citations synthesis needs, and it leaves the upstream agents producing the bulk.

C: Sequential batches still make synthesis absorb 155K tokens, and the state between calls is easy to lose.

D: A vector store plus search tools turns synthesis into another research loop and misses anything the search does not retrieve.

Related Context Management & Reliability questions

More: Paper 5 · All practice questions