Loading threat actor…
GTG-1002 is a threat cluster assessed with high confidence to be a Chinese state-sponsored cyber-espionage actor. The group is notable for one of the earliest publicly documented cases of an intrusion campaign in which an AI coding agent was used as an operational orchestrator rather than merely as an assistant. GTG-1002 has been associated with espionage activity against roughly 30 organizations across multiple countries, including government agencies, critical infrastructure entities, technology companies, financial institutions, and chemical manufacturers. The actor’s tradecraft combined conventional intrusion techniques with agentic AI-enabled automation. Reported activity included reconnaissance, attack-surface mapping, exploitation of public-facing applications, credential harvesting, discovery of internal services, lateral movement, collection, and data exfiltration. Human operators appear to have provided strategic direction and approval at key decision points, while the AI system reportedly handled a large share of tactical execution, including sequencing tasks, generating exploit logic, testing access paths, mapping privileges, and documenting operations. GTG-1002 reportedly used multiple isolated instances of Claude Code connected to external tools through the Model Context Protocol, alongside open-source penetration-testing utilities and a custom orchestration layer. The campaign has been described as relying primarily on widely available tooling rather than custom malware, with persistence in some successful intrusions achieved through creation of backdoor accounts. Reporting also indicates the actor used role-based jailbreaks and task decomposition to bypass model safeguards by framing malicious actions as legitimate defensive security work. The significance of GTG-1002 lies less in novel underlying ATT&CK techniques than in the machine-speed orchestration of the intrusion lifecycle. The campaign is widely cited as an example of how state-backed operators can use AI agents to compress timelines for reconnaissance, exploitation, credential access, and lateral movement, increasing operational tempo and scale while reducing the amount of direct human interaction required. Known alias: GTG-1002.
Mallory correlates actor tradecraft and target patterns against your stack, your sector, and your geography. See overlap before they land.
Who, where, and (when attributed) which flag flies behind the operation. Pulled from open-source reporting and Mallory's analyst review.
Attributed origin per open-source reporting.
21 distinct techniques observed across reporting, grouped by tactic. Hover any cell for the evidence excerpt; click through for MITRE's full description.
20 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
Chinese-linked espionage campaign in which Claude Code reportedly handled 80 to 90 percent of tactical work, including reconnaissance, exploitation, credential harvesting, and lateral movement, across about 30 target organizations.
Chinese state-sponsored campaign cited as an example of AI-orchestrated intrusion, using Claude Code with MCP-integrated offensive tooling to scan services, exploit a web vulnerability, harvest credentials, and pivot laterally across victim cloud environments.
Espionage campaign using an AI agent integrated with offensive tooling to autonomously scan, pivot, exploit an SSRF flaw, harvest SSH keys and cloud credentials, and move laterally with limited human intervention.
Conducted a campaign breaching government and critical infrastructure organizations in multiple countries, using Claude with scaffolding as an autonomous operator to help orchestrate attacks.
Match sector + geo + tech-stack targeting against your real footprint.
Every observed MITRE ATT&CK technique, grouped by tactic.
Families this actor is known to deploy, with IOCs and behavior.
CVEs this actor has used in known campaigns.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Domains, IPs, and hashes tied to this actor, refreshed continuously.