Research and help rebook a flight with an always-on agent
Last week, between back-to-back model reviews, company strategy calls, writing, and my never-ending hunt for apartment furniture on Facebook Marketplace, I tried to change a flight out of San Francisco. Task: The user attempted to rebook a flight out of San Francisco using OpenAI's Dot, an always-on agent. Dot researched flight options and cross-referenced Slack discussions to identify a potential scheduling conflict with a colleague's video recording, alerting the user despite the event not being on their calendar. The human role involved initiating the request and reviewing the agent's contextual warnings before proceeding.; tools: Dots. Human role: The original post does not say how a person stayed involved.; limitations: The agent requires access to communication platforms like Slack to detect conflicts not present in the calendar. The source is a paid subscriber article, limiting full context. Human approval is critical before any booking action, as the agent only provided a warning, not a completed transaction..
Tools
Published on Agent Work Atlas:
Overview
Last week, between back-to-back model reviews, company strategy calls, writing, and my never-ending hunt for apartment furniture on Facebook Marketplace, I tried to change a flight out of San Francisco.
Workflow
The user attempted to rebook a flight out of San Francisco using OpenAI's Dot, an always-on agent. Dot researched flight options and cross-referenced Slack discussions to identify a potential scheduling conflict with a colleague's video recording, alerting the user despite the event not being on their calendar. The human role involved initiating the request and reviewing the agent's contextual warnings before proceeding.
Reported result
The source does not report the final booking outcome, cost savings, or time saved. It only confirms that the agent successfully identified a hidden scheduling conflict from Slack data. A measurable acceptance check would be verifying that the agent correctly flags conflicts in 90% of test scenarios without false positives.
Human role
The original post does not say how a person stayed involved.
Limitations
The agent requires access to communication platforms like Slack to detect conflicts not present in the calendar. The source is a paid subscriber article, limiting full context. Human approval is critical before any booking action, as the agent only provided a warning, not a completed transaction.
Try it
Act as my travel assistant using OpenAI's Dot. I need to change my flight from [Origin City] to [Destination City] on [Date]. First, research available flights and compare prices. Second, scan my [Communication Platform, e.g., Slack] for any recent discussions about team meetings or events on [Date] that might conflict with my travel. If a conflict is found, alert me with the specific context. Do not book anything yet. Present the flight options and any conflict warnings for my review. I will confirm the final choice before you proceed with any booking actions. If no conflicts are found, list the top three flight options with prices and times.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.