Washington, November 17th / PRNewswire / -- Aimed at the " agentic workforce " proxy detection and response platform, Evoke Security today launches Evoke Defender Capture the Flag ( CTF ). This is a practical laboratory that helps security professionals investigate proxy-related incidents involving AI. Participants will handle five phased incidents, modeled after the proxy behaviors observed in real environments by Evoke. Analysts can clearly see what is required to investigate a proxy-related incident, as well as what will be missed without the correct visibility.
Evoke believes that the next incident most likely to be investigated by the security team will involve AI proxies. Nowadays, these proxies are used to execute commands on behalf of employees, read and transfer sensitive data, and perform operations in terminals, SaaS, and cloud systems. Most security teams have never investigated such incidents before and are not clear about which systems they need to preserve evidence. Defender CTF advises them to figure this out now in the laboratory, rather than facing it only when an intrusion occurs.
In the laboratory, participants will receive a complete record of all operations performed by the agent. These records are derived from similar data collected by Evoke in the customer environment: prompts and instructions received by the agent, each call to the MCP server and its input content, commands executed on the terminal, accessed files and records, as well as actions taken within the SaaS application and cloud services. Each item can be traced back to the relevant person and agent. Participants will be able to see the source of each piece of evidence, what it reveals, and how this information fits together to form a complete timeline.
This CTF focuses on "out-of-control" agents, that is, agents who deviate from their assigned tasks and act beyond their boundaries. Each step may seem reasonable on its own, but CTF demonstrates how these steps can accumulate into a security incident, and what is required to prove this.
When handling cases, participants need to distinguish between what the user requested and what the agent actually did, identify when the agent exceeded instructions, mark suspicious behavior, and reconstruct the timeline of the events. They will also realize how much information a case would lose without agent telemetry. Terminals, SIEM, and identity logs can show that a certain process was running or that a file was accessed, but they cannot reveal the underlying commands, the tool calls in between, or why the agent took that action.
“Every investigation depends on evidence. When it comes to agents, the most important evidence—what they were asked to do, what decisions they made, which tools they used—is not found in the logs that most teams already have,” said Evoke Security, co-founder of Jason Rebholz. “We created Defender CTF so that practitioners can see this evidence for themselves, and to understand what an agent investigation would look like when such evidence is not available. Without this data, it is simply impossible to conduct such investigations.”
CTF demonstrates the full depth of the content captured by Evoke, but in daily use, the design goal of Evoke is to minimize the need for the security team to manually handle proxy cases. The platform continuously monitors proxies, performs threat modeling on them, and blocks high-risk behaviors in real-time, while also providing complete evidence records when investigations are necessary.
Defender CTF is designed for incident responders, SOC analysts, threat hunters, and detection engineers, and was created by the incident response teams of Evoke. These team members possess intrusion investigation experience from Mandiant. Evoke indicates that it is not yet known if there are any other practical exercises built around the AI proxy incidents. Upon completion, participants will receive a useful map regarding the sources of proxy evidence and how it was used in the investigations, as well as a clear understanding of the visibility gaps in their own environments.
"Use your investigative skills to find out what is needed to investigate this AI proxy incident. Visit Evoke Defender CTF."
About Evoke Security
Evoke Security is a proxy detection and response platform designed for agent-based employees, focusing on security protection. The company was founded by seasoned professionals in incident response, aiming to assist security teams in continuously inventorying, threat modeling, and monitoring AI proxies, thereby preventing high-risk proxy activities in real-time. Evoke views proxies as the next generation of operating systems for enterprises, helping organizations adopt them without further accumulating security technology debt.
Correction:The first paragraph has been updated.












