The China-aligned Advanced Persistent Threat (APT) group known as Webworm (also tracked as Space Pirates) has significantly evolved its tactics and expanded its targeting scope from Asia to European government entities. Research from ESET reveals the group is deploying a new arsenal of sophisticated backdoors designed for stealth and persistence. Two notable new malware families, EchoCreep and GraphWorm, abuse legitimate, high-reputation cloud services for command-and-control (C2) communications. EchoCreep uses Discord's API, while GraphWorm leverages the Microsoft Graph API and OneDrive. This living-off-the-trusted-land (LOTL) approach makes their C2 traffic exceptionally difficult to detect, as it blends in with legitimate business communications. The group's primary objective appears to be cyber-espionage, with observed exfiltration of sensitive government documents from targets in Spain, Belgium, Italy, and Poland.
Webworm is a sophisticated espionage group that continuously refines its toolset to evade detection. Their latest campaign demonstrates a clear focus on stealth and operational security.
discord.com, graph.microsoft.com). This bypasses many network security controls that rely on domain reputation and blacklisting.WormFrp, to tunnel their traffic and further obscure their activities.The attack chain showcases the group's methodical approach:
T1566.001 - Spearphishing Attachment) or exploit public-facing applications (T1190 - Exploit Public-Facing Application).T1071.001 - Application Layer Protocol: Web Protocols: The core of the C2, using legitimate web services.T1105 - Ingress Tool Transfer: Downloading the backdoors after initial access.T1567.002 - Exfiltration to Cloud Storage: Using OneDrive and AWS S3 for data exfiltration.T1573.002 - Asymmetric Cryptography: Using standard TLS encryption provided by Discord and Microsoft to hide C2 traffic.T1027 - Obfuscated Files or Information: Encrypting the C2 messages sent via Discord.No specific IOCs such as domains, IPs, or file hashes were provided in the summarized articles.
Security teams should hunt for anomalous use of legitimate services:
url_patterndiscord.com/api/webhooks/url_patterngraph.microsoft.com/v1.0/me/drive/process_namesvchost.exenetwork_traffic_patternUnusual User-Agent for graph.microsoft.comlog_sourceMicrosoft 365 Audit Logspowershell.exe or a random executable connecting to graph.microsoft.com is a major red flag.China-aligned Webworm APT now widely using SoftEther VPN for persistent, encrypted C2 channels, expanding evasion tactics.
Implement strict egress filtering to block access to unauthorized web services like Discord from server environments.
Use application-aware firewalls or CASB solutions to block access to personal cloud storage and other services not sanctioned by the organization.
Deploy EDR solutions capable of detecting malicious processes and their anomalous network connections, even when they are to trusted domains.
Outbound Traffic Filtering is the most effective defense against Webworm's C2 tactics. The backdoors rely on connecting to public services like discord.com and graph.microsoft.com. Organizations must implement a default-deny egress policy on servers. Connections to Discord should be blocked entirely from server subnets, as there is no legitimate business reason for them. For Microsoft Graph, a more nuanced approach is needed. Use an application-aware firewall or proxy that can enforce Microsoft 365 tenant restrictions. This ensures that even if a process tries to connect to the Graph API, it can only authenticate to the organization's own tenant, not a personal or attacker-controlled one. This filtering breaks the C2 channel, rendering the EchoCreep and GraphWorm backdoors inert.
Process Analysis via an EDR is crucial for detecting the backdoors themselves. Security teams should create detection rules that hunt for suspicious process and network event correlations. For example, a rule should alert when a non-browser process (like svchost.exe, rundll32.exe, or an unknown executable) initiates a network connection to discord.com or graph.microsoft.com. Baselining normal network activity for legitimate processes is key. The EDR should also monitor for process injection or the loading of suspicious modules that might contain the backdoor's logic. This allows for detection of the malware's execution on the endpoint, even if the network traffic itself is encrypted and destined for a trusted domain.
For the GraphWorm backdoor, which abuses a Microsoft account, Domain Account Monitoring (extended to cloud identities) is a vital detection layer. Ingest Microsoft 365 Unified Audit Logs into a SIEM. Hunt for anomalous activity associated with user or service accounts. Key indicators for GraphWorm would be: 1) An account accessing OneDrive via the Graph API from an unusual IP or location. 2) A pattern of small file uploads (tasking) followed by larger file uploads (exfiltration) in a specific OneDrive folder. 3) API-only access to a OneDrive account that is normally accessed interactively by a user. By monitoring the behavior of the cloud identity, defenders can spot the abuse of legitimate credentials and infrastructure, revealing the C2 dead-drop.
Webworm operators observed exfiltrating files from a Spanish governmental entity between December 2025 and January 2026.
ESET publishes research detailing Webworm's new backdoors and expansion into Europe.

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.
Help others stay informed about cybersecurity threats
Every tactic, technique, and sub-technique used in this threat has been identified and mapped to the MITRE ATT&CK framework for consistent, actionable threat language.
Observables and indicators of compromise (IOCs) have been extracted and cataloged. Risk has been assessed and correlated with known threat actors and historical campaigns.
Detection rules, incident response steps, and D3FEND-aligned mitigation strategies are included so your team can act on this intelligence immediately.
Structured threat data is packaged as a STIX 2.1 bundle and can be visualized as an interactive graph — relationships between actors, malware, techniques, and indicators.
Sigma detection rules are derived from the threat techniques in this article and can be converted for deployment across any major SIEM or EDR platform.