Entity trail

FortiGate

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-03-27 / sources-researcherAstrix Security: Four-Method AI Agent Discovery Architecture + Real-Time Policy Engine at RSAC 2026Astrix launched a four-method agent discovery system at RSAC 2026: AI platform integrations (Microsoft Copilot, Bedrock, Vertex, OpenAI, Agentforce), NHI fingerprinting (detecting agents from OAuth apps, service accounts, API keys, PATs they use), EDR sensor telemetry (CrowdStrike, SentinelOne, Defender, FortiGate), and Bring-Your-Own-Service for proprietary tools. New Agent Policies feature evaluates allow/flag/block rules scoped by user, department, platform, and resource type before an action executes.
2026-03-07 / agents-researcherCyberStrikeAI: First AI-Native Offensive Tool Deployed at ScaleAn open-source AI-powered attack platform called CyberStrikeAI was used to compromise 600+ FortiGate devices across 55 countries between January-February 2026. Developed by a China-based individual with ties to CNNVD (Chinese National Vulnerability Database), 21 unique IPs were observed running the platform in just 37 days. The tool exploited exposed management ports and weak credentials — not zero-days — but its rapid adoption demonstrates that AI-native offensive tooling has crossed from proof
2026-03-07 / news-researcherCyberStrikeAI: Open-Source AI Attack Platform Compromises 600+ FortiGate Devices Across 55 CountriesAmazon Threat Intelligence discovered a Russian-speaking financially motivated actor using CyberStrikeAI — an open-source Go-based offensive tool maintained by Chinese developer Ed1s0nZ — to compromise 600+ FortiGate devices across 55 countries between January 11 and February 18, 2026. The tool integrates 100+ security tools and leveraged Anthropic Claude and DeepSeek for AI-powered vulnerability discovery. Team Cymru identified 21 unique CyberStrikeAI IPs hosted in China, Singapore, and Hong Ko

Graph relationships

MENTIONS
Amazon -> FortiGate
Amazon Threat Intelligence documented AI-assisted attacks compromising FortiGate devices.
Source finding
MENTIONS
CyberStrikeAI -> FortiGate
CyberStrikeAI was used to compromise 600+ FortiGate devices across 55 countries.
Source finding
MENTIONS
CyberStrikeAI -> FortiGate
CyberStrikeAI compromised 600+ FortiGate devices across 55 countries.
Source finding

Source trail

The Hacker News Deep-dive analysis HexStrike framework zero authentication AWS Security Blog The Hacker News BleepingComputer — FortiGate

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues