ThreatBook Launches RedAgents for Parallel AI Pentesting
The new SaaS platform allows security teams to run multiple penetration testing agents simultaneously via a unified interface.
Updated on Oct. 16, 2026 in Artificial Intelligence
Live Poll
Do you believe using multiple AI security tools provides better protection than using one vendor?
ThreatBook has launched RedAgents, a software-as-a-service platform that aggregates multiple AI penetration testing agents. The tool enables users to run several security testing agents in parallel through a unified interface.
Why it matters
RedAgents seeks to eliminate fragmented security monitoring by integrating disparate agents into a single workflow. This approach aims to improve vulnerability detection rates for enterprise security teams.
The platform builds on the capabilities of specialized agents, such as Cyber(YanJi), which achieved a perfect score of 54/54 problems in recent competitive testing. The system relies on centralized management to coordinate simultaneous penetration testing tasks within defined whitelist parameters.
The players
ThreatBook
A cybersecurity company with operational hubs in Singapore, Seoul, and Palo Alto that provides AI-powered security infrastructure.
CyberStrikeAI
A security research organization acquired by ThreatBook that previously garnered 7,800 GitHub stars for its open-source contributions.
Cyber(YanJi)
An AI penetration testing model that achieved a perfect score at the Tencent Cloud Hackathon.
The details
RedAgents functions by aggregating multiple AI penetration testing agents into a centralized, unified management dashboard. Users define the testing scope by configuring a whitelist within authorized parameters, allowing the agents to conduct parallel security assessments. This architecture is designed to synthesize results from diverse agents, mitigating the limitations of traditional, siloed security monitoring systems.
Timeline
September 3, 2026: Artex (Open-Source Edition) won Baidu's Agent+ Offensive and Defensive Challenge.
September 2026: ThreatBook acquired CyberStrikeAI.
October 8, 2026: Artex (Open-Source Edition) transitioned to a closed-source model.
October 16, 2026: ThreatBook launched the RedAgents platform.
The Tech Race
RedAgents represents a shift from independent, open-source security models toward integrated, closed-source commercial platforms. This follows the recent acquisition of CyberStrikeAI and the pivot of the Artex project from a community-driven open-source utility to a proprietary enterprise solution.
Security teams can utilize the platform to replace manual, siloed agent deployments with a coordinated, parallelized testing workflow. Users interested in the platform can apply for access or product information through the provider's portal.
The takeaway
The platform signals an industry trend toward consolidating specialized security agents into unified management suites. Security teams should monitor whether this aggregation strategy improves mean time to detection compared to current standalone testing tools.
Further reading
For broader trends in secure AI development, see our coverage of Artificial Intelligence.
More information
For complete platform details, visit the RedAgents product information and trial application.
Source note: This article includes information reported by The Queenslander.
Live Poll
Do you believe using multiple AI security tools provides better protection than using one vendor?
