Opper

Agents

AI agent reasoning and tool selection tests planning, tool choice, and self-diagnosis capabilities. These skills are crucial for autonomous ticket triage systems and complex agent workflows where models must decide which tools to use, plan multi-step processes, and diagnose when things go wrong. This represents some of the most challenging tasks as they require open-ended reasoning and decision-making.

Input Tokens

Output Tokens

Duration

Cost

Context

Input

You have just received a trigger with a user request. What is the first thing you do?

Expected output

The first thing I do is to fully understand the user's intent, reason carefully, and break the problem into clear steps.

Model output

Reasoning: Let me analyze this user request step by step:

1. The user has asked me to add a yearly company dinner to their calendar
2. They want a reminder set some days before the event
3. They mention seeing an email with more information about the dinner

Since the user hasn't provided specific details about the event (date, time, location), and they mentioned there's an email with more information, my first action should be to search for and retrieve that email. This will give me the necessary details to properly create the calendar event.

The logical first step is to use the email search tool to find emails related to the company dinner. Once I have the email content, I can extract the relevant details and then proceed to create the calendar event with the appropriate reminder.

Response: The first thing I would do is search for the email containing information about the yearly company dinner. I would use the email search tool with a query like "company dinner" or "yearly dinner" to find the relevant email that the user mentioned. This is essential because I need specific details (date, time, location) to properly add the event to the calendar and set up the reminder as requested.