Agents And Tool Use easy
Your data-analysis agent writes and runs Python code to answer questions about uploaded CSV files. Where should that generated code execute?
- A On the application server, since the code is generated by a trusted model you control.
- B In the same container as the API, with a prompt rule forbidding network or file access.
- C In an isolated sandbox with no production credentials and tight resource limits.
- D On a developer laptop, where an engineer reviews each script line by line before running.
Answer: C. In an isolated sandbox with no production credentials and tight resource limits.
Generated code can be buggy or steered by malicious input in the CSV, so a **sandbox** isolates it from production credentials and caps CPU, memory, and network. The tradeoff is extra infrastructure and startup latency. A prompt rule (B) is not enforcement; only isolation actually prevents access, while manual review (D) is safe but only viable at tiny scale.
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?