Agents And Tool Use easy
During testing, an agent gets stuck alternating between two search tools and runs for 200 steps before someone kills it. What is the most direct fix?
- A Add more tools so the agent has better options when its current tools stop being useful.
- B Enforce a hard cap on steps, tokens, and wall time, and end the run when one is exceeded.
- C Ask the model in the prompt to stop as soon as it believes the task has been completed.
- D Switch to a larger context window so the agent can see its full history and notice loops.
Answer: B. Enforce a hard cap on steps, tokens, and wall time, and end the run when one is exceeded.
**Budgets** enforced by the host (commonly 10 to 30 steps for a task agent) guarantee runaway loops terminate, trading off the occasional legitimate long task that gets cut short. A prompt-level stop rule (C) is a useful hint for normal completion, but it cannot be relied on when the model is the thing that's looping.
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?