Agents And Tool Use easy
An agent's search tool returns full web pages of about 30,000 tokens each. After a few steps, cost and latency climb sharply. What is the best fix?
- A Move to a model with a larger context window so the full pages always fit in the prompt.
- B Reduce the step cap to two so the agent cannot accumulate more than two large results.
- C Cache the model's final answers so repeated questions skip the tool calls entirely.
- D Truncate or summarize tool results to a few thousand tokens before adding them back.
Answer: D. Truncate or summarize tool results to a few thousand tokens before adding them back.
Each loop step re-sends the whole growing transcript, so cost grows roughly quadratically with steps when tool results are huge. **Truncating or summarizing** results trades some detail for much cheaper steps. A larger context window (A) makes the pages fit but does nothing about paying for them again on every step.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You are defining a 'schedule_meeting' tool for an agent. Which schema decision most reduces wrong or malformed calls?
- A model emits three parallel tool calls: `get_inventory`, `get_price`, and `get_shipping`. To cut latency, the host appends each result to the transcript as it arrives and immediately re-invokes the model. Runs start failing with malformed-request errors, and some produce hallucinated shipping costs. What should the host do?
- A team splits a coding agent into planner, coder, tester, and reviewer agents, and each receives the full transcript of the previous agent. The expected quality gains did not appear. Cost tripled, and the reviewer often fixates on approaches that earlier agents abandoned. When is a multi-agent split actually justified?
- A support agent should remember each customer's preferences across months of sessions. The team appends every past session summary to the system prompt. After six months, prompts exceed 80,000 tokens, latency has doubled, and the agent sometimes ships to an address the customer replaced weeks ago. What redesign fixes both problems?
- A teammate proposes splitting a single coding agent into five agents, arguing that more agents means smarter results. When is a multi-agent split actually justified?
- A data-analysis agent runs model-generated Python inside the same container as your API service. That container has database admin credentials in environment variables. A security review flags this as critical. What is the best mitigation?