Generating plain-English description…
Plain-English summary of the active rules — what agents can do, what gets blocked, and what needs a human.
This policy applies only to the **research_crew** agent in the **dev** environment. It may use only the **arxiv_search**, **web_search**, and **delete_research_data** tools, and only the **gpt-4o-mini** model; however, **delete_research_data** requires approval before it can run. Requests are limited to **4,000 tokens per request**, with usage warnings at **80%**, and overages are tightly controlled: repeated over-limit use is denied after **3** overages in a **24-hour UTC** window, with a hard deny at **2×** the allowed overage threshold. Safety guardrails require **PII to be redacted**, and any **prompt injection** or **data exfiltration** attempt must be **denied**.