Introducing Scheduled Prompts: Automated Security Reports with AI

Today RunReveal is announcing Scheduled Prompts and our customers can begin using this feature immediately.

We couldn’t be more excited to release this feature and we are confident it will help our customers better understand their security posture using agentic AI.

What are scheduled prompts?

Scheduled Prompts allow our customers to prompt RunReveal’s AI agent using a configurable prompt at a scheduled time. The scheduling is configured with cron, and the results are delivered to a notification channel of your choosing in markdown format.

Since we released the RunReveal MCP server a few months back, and our own AI agent last month, we’ve seen first hand how powerful AI tools can be in summarizing, querying, and providing visibility into your security data.

We’re incredibly excited to provide our customers with the same amazing experience that our MCP server and RunReveal chat experience provides, but without needing to have a human sitting in front of a keyboard building out complicated queries. 

Demo

To kick off a Scheduled Prompt you’ll need to decide on a schedule that you want to run the agent, and configure a prompt that gives the agent its task. One obvious first thing to try is summarizing the previous day’s alerts and findings. In this case below we want a report delivered to our inbox at 7am UTC time on Monday through Friday. We also can additionally specify which AI provider to use for this.

At the scheduled time the AI agent will be invoked, along with all of the RunReveal tools, and we’ll send the results when it completes via markdown to a configured notification channel. Here’s an example of the results from the above prompt.

The use cases for this type of technology is only limited by your imagination, and we expect our customers will use this type of thing for:

• Monitoring high risk actors in their organization
• Aggregating information about specific types of threats on a regular basis
• Monitoring for changes in the quantity of alerts they are receiving and the classification of those alerts.
• So much more!

Any type of question you might be able to answer by querying your logs can now be delivered to your email inbox with just a few keystrokes.

What’s next

We plan to improve this product over the coming months. This is the first “headless” invocation of the RunReveal agent, but there’s more possibilities for integrating it into our product. We will likely add support for automatic investigations of specific alerts in Q4 of this year.

This blog post is part of our #RunRevealWeek where our team is sharing product announcements, behind the scenes on how we approach problems, and the unique features that make us, us.