ThreatBook Launches RedAgents AI Platform That Uses Multiple Hacking Agents to Find Security Flaws

ThreatBook has launched RedAgents, a cloud platform that brings several AI-powered penetration-testing agents together to assess an organization’s systems.

Announced on October 16, 2026, the service is designed to help security teams find weaknesses in their own networks by running multiple automated tests within a defined, authorized scope.

ThreatBook says the agents can work in parallel and show their findings and test logs in one place. The platform takes a different approach from relying on a single testing tool or provider.

Each agent may look for different issues, so combining their results could give defenders a broader view of their exposure.

The launch announcement describes RedAgents as a software-as-a-service platform for security operations teams, rather than a new malware family or a report on a specific intrusion.

ThreatBook’s announcement also notes that the service brings together agents with different origins and capabilities. Those named at launch are Artex Pro, Cyber(YanJi), Yanxun and CyberStrikeAI Enterprise Edition.

The company says teams can choose one agent or run several against the same approved target, then review results through a shared interface.

ThreatBook Launches RedAgents AI Platform

RedAgents acts as an aggregator: it connects multiple AI penetration-testing agents through one platform and can run them in parallel. A penetration test is an authorized review in which testers look for weaknesses in systems before malicious actors can exploit them.

Running agents side by side may help uncover issues that a single tool misses, though the value of the results depends on how accurately the agents test and report.

ThreatBook says the platform consolidates findings, test data and logs in a unified view. This could reduce the need for teams to move information between separate tools.

The approach is similar to using multiple scanners or testing teams, but RedAgents is designed to coordinate AI agents as part of one assessment.

For readers comparing different designs, Cybersecurity News has covered a multi-agent pentesting toolkit that assigns different testing tasks to specialist agents.

The four named agents have distinct backgrounds. ThreatBook describes Artex Pro as the enterprise version of Artex, an autonomous penetration-testing agent. Cyber(YanJi) is a Python-based agent that searches for a route from a starting point to a defined goal.

Yanxun is described as a system that coordinates specialist agents, while CyberStrikeAI Enterprise Edition is based on an AI-native testing tool that ThreatBook acquired in September 2026.

These descriptions come from the company’s announcement and should be read as vendor claims, not as independent comparative test results.

ThreatBook reports, for example, that Yanxun can coordinate several dozen agents and that Cyber(YanJi) solved 54 of 54 challenge problems during a validation event.

The release does not provide a controlled, side-by-side benchmark showing how all four agents perform against the same targets.

Scope and data controls

Automated testing tools can cause disruption if they probe systems without permission or exceed an agreed scope. ThreatBook says RedAgents lets customers define an allowlist of approved targets and that agents must operate within those authorized limits.

Organizations should still confirm the scope, timing, rate limits and test conditions before running any penetration test, especially on systems that support business operations.

The company also says customer test data stays within the ThreatBook environment and is not sent to third-party agents. RedAgents supports Zero Data Retention, or ZDR, which ThreatBook says enables complete deletion of test data.

Security teams should confirm how these controls work in practice, including retention periods, access permissions, logging and any contractual terms, before submitting sensitive environment details.

The launch comes as more security products use AI agents to automate parts of testing. Other examples covered by Cybersecurity News include PentestCode’s AI testing agent and Xalgorix’s verification-focused platform.

These products differ in how they are deployed and how they handle findings, so teams should compare scope controls, evidence quality and data handling rather than treating “AI pentesting” as one uniform category.

The main promise of RedAgents is breadth: multiple agents may explore a target in different ways and give teams a combined view of possible weaknesses.

That does not guarantee that every result is valid or that every vulnerability will be found. Security staff still need to review evidence, reproduce important findings safely and prioritize fixes based on business impact and exposure.

ThreatBook says the platform can reduce false positives by combining agent results, but the announcement does not publish independent accuracy figures or a false-positive rate.

It also does not identify a specific customer deployment, measured security outcome or external validation of the service. Those details will matter to organizations assessing whether the platform fits their testing process.

For context, a separate Cybersecurity News report describes Apex’s black-box testing approach, which evaluates applications without source-code access.

RedAgents’ stated distinction is not a single testing method but a central platform for selecting and coordinating several agents. ThreatBook says it plans to add more open-source agents over time, alongside the enterprise tools available at launch.

The company’s announcement and RedAgents product blog provide further product details. RedAgents is available through the company’s website, where ThreatBook says organizations can request a trial.

Indicators of compromise (IoCs):-

IoC type	Value
Malware name	Not reported
File hash / SHA value	Not reported
Domain or IP address	Not reported
Other indicators	Not reported

Note: IP addresses and domains are intentionally defanged (e.g., [.]) to prevent accidental resolution or hyperlinking. Re-fang only within controlled threat intelligence platforms such as MISP, VirusTotal, or your SIEM.

Update defenses against new malware as soon as it appears. Integrate TI Feeds with data from 16K SOCs.

The post ThreatBook Launches RedAgents AI Platform That Uses Multiple Hacking Agents to Find Security Flaws appeared first on Cyber Security News.