{
	"id": "81815d75-a4ff-460b-abe5-04c691ad1992",
	"created_at": "2026-11-17T02:11:39.14589Z",
	"updated_at": "2026-11-17T02:19:45.951198Z",
	"deleted_at": null,
	"sha1_hash": "06413330a7f7a014d4da313667428546df094103",
	"title": "",
	"llm_title": "",
	"authors": "Federal Bureau of Investigation",
	"file_creation_date": "2026-11-13T12:57:06Z",
	"file_modification_date": "2026-11-13T12:57:06Z",
	"file_size": 541176,
	"plain_text": "TLP:CLEAR\n\n13 NOVEMBER 2026\n\nFLASH Number \nFLASH-20261113-001\n\nThreat Overview of ARTEX Autonomous Penetration\n\nTesting Framework\n\nSummary\n\nThe Federal Bureau of Investigation (FBI) is releasing this FLASH to highlight the tactics, techniques, and \nprocedures (TTPs) and indicators of compromise (IOCs) associated with the ARTEX autonomous \npenetration testing framework. ARTEX is an open-source, large language model (LLM)-driven agentic \nplatform that threat actors have weaponized to automate multiple phases of cyber intrusions at machine \nspeed, from reconnaissance through data exfiltration. Since at least August 2026, a variety of malicious \ncyber actors have used ARTEX to conduct campaigns targeting organizations globally across multiple \nsectors, including financial services, government, technology, defense, semiconductor, energy, \nhealthcare, manufacturing, telecommunications, and legal industries.1,2,3,4,5\n\nARTEX significantly reduces the time required to chain attack techniques, with some intrusions \nprogressing from initial targeting to privileged access in under three hours.² The framework enables threat \nactors with minimal technical expertise to conduct complex, multi-target operations simultaneously, \nfundamentally accelerating attack velocity and expanding operational scale. Notable campaigns include \nfinancially motivated attacks against South Korean financial institutions resulting in exfiltration of personal \ninformation from approximately 67,000 individuals,6,7,8 and a supply chain compromise affecting \nConnectWise Cloud Backup customers that enabled access to more than 2,250 downstream tenants.2\n\nFollowing public disclosure of malicious usage, the original ARTEX developer removed the GitHub \nrepository in October 2026.3 However, the framework continues to proliferate through community forks, \nlocalized variations, archived copies, and a commercial version (ARTEX PRO), ensuring continued \navailability to threat actors.3,10 As of early November 2026, researchers identified over 1,600 publicly \nexposed ARTEX console instances, with total deployments likely significantly higher when including \nsystems behind firewalls or in private networks.4\n\nThe FBI encourages organizations—particularly those operating internet-exposed applications, self-\nservice portals, cloud infrastructure, and partner/supplier systems—to implement the recommendations \nin the Mitigations section to reduce the likelihood and impact of ARTEX-enabled intrusions.\n\nTLP:CLEAR\n\nTLP:CLEAR\n\nThreat\n\nARTEX is an autonomous, multi-agent penetration testing platform written in Go.3,4 The framework \ntranslates high-level tactical goals described in natural language into actionable machine commands \nexecuted by specialized super intelligence (SI) agents, requiring minimal human interaction beyond initial \nobjective setting.2,3 ARTEX does not include its own SI model but integrates with external LLM providers, \nincluding frontier models (Anthropic Claude, OpenAI GPT) and open-source or Chinese-origin models \n(DeepSeek, GLM, Grok, Kimi).3,6,7,9\n\nAnalysis of attacker-controlled ARTEX instances shows Chinese models are most prevalent, with \nDeepSeek Flash family models dominating usage due to their balance of capabilities, relatively low cost, \nand lack of observable guardrails when framed for penetration testing.7,9\n\nOriginally released on GitHub by developer “Autumn-27” in July 2026, ARTEX gained rapid adoption, \nreaching an estimated 1,500–2,000 GitHub stars within 10 weeks.4 The developer stated the tool was \nintended for authorized security testing within defined scopes and made the repository unavailable on \nOctober 8, 2026, following reports of threat actor abuse.3,6 Despite this action, ARTEX remains widely \naccessible through independent routes and mirrors on code-sharing platforms.3,4,9\n\nObserved Capabilities and Potential Uses\n\nBased on observed capabilities and ongoing threat actor adoption, ARTEX presents the following risks:\n\n•  Accelerated Multi-Target Operations: SI agents provide automation for the time-consuming \nphases of enumeration and exploitation; small threat groups can run concurrent intrusion \ncampaigns at a scale previously unavailable to them. \n•  Exploitation of Overlooked Attack Surfaces: ARTEX excels at identifying and exploiting peripheral \nsystems receiving less security attention than flagship applications, including partner and supplier \nportals, mobile backends, employee applications, sales-support systems, legacy services, third-\nparty integrations, and applications created across different business units.1  \n•  Supply Chain Compromise: Agents have demonstrated capability to identify and exploit supply \nchain attack vectors, including modifying iOS application distribution infrastructure via exposed \ncloud credentials.9 Autonomous agents can systematically map trust relationships between \norganizations. By identifying and compromising weaker third-party vendors, CI/CD pipelines, or \nintegrated software providers, attackers can leverage ARTEX to pivot from a peripheral supplier \ndirectly into high-value upstream enterprise networks. \n•  Reduced Barrier to Entry: ARTEX lowers technical barriers for conducting complex intrusions, \nenabling threat actors with minimal expertise to execute sophisticated campaigns.3,9\n\nTLP:CLEAR\n\nTLP:CLEAR\n\n•  Compressed Detection Window: ARTEX generates high volumes of forensically noisy traffic that \nmay trigger alerts, but attack speed provides limited time for coordinated defender response \nbefore objective completion.2,3\n\nObserved Tactics, Techniques, and Procedures\n\nThe following table maps observed ARTEX activity to MITRE ATT\u0026CK phases:\n\nPhase \nInitial Recon \nObserved Activity \n• Automatically mapped \neffective permissions of Azure \ncredentials (SAS tokens, \nService Principals, Kudu \ncredentials, Storage Account \nkeys) using sub-second \ncapability checks, test blob \nwrites/deletes, ARM \ndeployment validation dry-\nruns \n• Tested compromised IAM \nuser permissions using \nDryRun flags \n• Fuzzing web directories \nacross 500+ requests \n• WHOIS asset mapping and \nactive vulnerability scanning \n• Queried api.db-ip[.]com to \ndetermine victim geolocation \nInitial Access  • Exploited unsigned SAML \nassertions \n• Exploited XXE flaws \n• Exploited anonymous file \nupload \n• Abused cross-account cloud  T1195 – Supply Chain \ntrust relationships from an \nexternal partner tenant \n• Compromised third-party \nvendor credentials to pivot via \ntrusted B2B federation \n• Exploited SSRF \n• Hardcoded administrator\n\nTLP:CLEAR\n\nATT\u0026CK Techniques \nT1595 – Active Scanning \nT1580 – Cloud Infrastructure \nDiscovery \nT1087.004 – Account \nDiscovery: Cloud Account \nT1069.003 – Permission \nGroups Discovery: Cloud \nGroups  \nT1526 – Cloud Service \nDiscovery\n\nT1190 – Exploit Public-Facing \nApplication \nT1078 – Valid Accounts \nT1133 – External Remote \nServices\n\nCompromise \nT1199 – Trusted Relationship \nT1606.002 – Forge Web \nCredentials: SAML Tokens \nT1552.001 – Unsecured \nCredentials: Credentials in \nFiles\n\nEstablish Foothold / \nPersistence\n\nPrivilege Escalation\n\nTLP:CLEAR\n\npasswords in SPA JavaScript \nbundles \n• Missing authentication on \nWeb APIs \n• RCE through command \ninjection \n• Insecure Direct Object \nReferences (IDOR) \n• Weak administrative \ncredentials \n• RCE through SSTI \n• SQL injection \n• Authentication bypass \n• Used stolen VPN \nconfigurations and leaked \ncredentials \n• Deployed JSP and PHP web \nshells (including via MySQL \nlog poisoning) \n• Abused Kudu administrative \nAPIs to execute arbitrary \ncommands \n• Established C2 on internal \nWindows database servers \n• Installed open-source RATs \n(WISP, vshell, AdaptixC2) \n• Created SSH key pairs, \nallocated Elastic IPs \n• Account takeover (password \nresets) \n• Added credentials to existing \nOAuth applications \n• Launched cloud compute \ninstances with high-value IAM \nroles \n• Kerberoasting to harvest \nservice tickets \n• SQL injection to dump \ndatabase credentials \n• DCSync to steal domain \nkrbtgt key \n• Created accounts with\n\nTLP:CLEAR\n\nT1505.003 – Server Software \nComponent: Web Shell \nT1078.004 – Valid Accounts: \nCloud Accounts \nT1098.001 – Account \nManipulation: Additional \nCloud Credentials \nT1574.001 – Hijack Execution \nFlow: DLL \nT1036.005 – Masquerading \nT1105 – Ingress Tool Transfer \nT1059 – Command and \nScripting Interpreter\n\nT1134.001 – Access Token \nManipulation \nT1098.003 – Account \nManipulation: Additional \nCloud Roles \nT1552.005 – Unsecured \nCredentials: Cloud Instance \nMetadata API \nT1606.002 – Forge Web \nCredentials: SAML Tokens\n\nInternal Recon\n\nLateral Movement\n\nCollection / Exfiltration Prep\n\nTLP:CLEAR\n\nis_admin=true flag \n• Abused resource-based \nconstrained delegation \n• Queried cloud metadata \nservice via XXE \n• Forged administrator token \nusing unrotated \nauthentication keys \n• Ran iterative PowerShell \nenumeration scripts against \nDomain Controllers \n• Enumerated over 110,000 \nActive Directory objects \n• Read LAPS passwords \n• Inventoried endpoint sensor \ncoverage to identify Windows \nmachines without \nCrowdStrike Falcon\n\nto domain controllers using \nPass-the-Ticket, WinRM,\n\n(blocked by endpoint \nprotection)\n\n• Interrogated backend SQL \ndatabases \n• Mass blob reads \n• Deployed custom boot \nscripts via EC2 instances to \nharvest data \n• Sent command output to \nrotating webhook.site \nendpoints \n• Read mailboxes on\n\nTLP:CLEAR\n\n• Attempted lateral movement  T1550.003 – Use Alternate\n\nKerberos relay, Pass-the-Hash  T1550.002 – Use Alternate\n\nT1606 – Forge Web \nCredentials \nT1190 – Exploit Public-Facing \nApplication \nT11558.003 – Steal or Forge \nKerberos Tickets: \nKerberoasting\n\nT1518.001 – Software \nDiscovery: Security Software \nDiscovery \nT1555 – Credentials from \nPassword Stores \nT1087.002 – Account \nDiscovery: Domain Account \nT1552.006 – Group Policy \nPreferences \nT1059.001 – Command and \nScripting Interpreter: \nPowerShell\n\nAuthentication Material: Pass \nthe Ticket\n\nAuthentication Material: Pass \nthe Hash \nT1021.006 – Remote Services: \nWindows Remote \nManagement (WinRM) \nT1557.001 – Adversary-in-the-\nMiddle: LLMNR/NBT-NS \nPoisoning and SMB Relay  \nT1114.002 – Email Collection: \nRemote Email Collection \nT1555.006 – Cloud Secrets \nManagement Stores \nT1530 – Data from Cloud \nStorage \nT1005 – Data from Local \nSystem \nT1074.001 – Data Staged: \nLocal Data Staging\n\nTLP:CLEAR\n\ncompromised mail server \n• Created new 100GB volumes  Remote Data Staging \nfor exfiltration preparation\n\nCommand and Control  • Used residential proxy \nservices \n• Routed traffic through Tor, \nSOCKS proxies, compromised  T1572 – Protocol Tunneling\n\nIndicators\n\nrelays \n• Tunneled traffic via Chisel \nand SoftEther \n• BLOODALCHEMY \ncommunicated with C2 \ndomains\n\nT1074.002 – Data Staged:\n\nT1059 – Command and \nScripting Interpreter \n \nT1090 – Proxy \nT1090.003 – Proxy: Multi-hop \nProxy\n\nT1071.001 – Application Layer \nProtocol: Web Protocols\n\nThe FBI recommends organizations investigate or vet IOCs prior to taking action, such as blocking. The FBI \nrecognizes that infrastructure used for malicious activity may be hosted on cloud or virtualized platforms \nwhere IP addresses can be dynamically or temporarily assigned. As a result, an address associated with \nmalicious activity at one point in time may later be reassigned to benign services. Accordingly, these \nindicators should be interpreted as historically observed infrastructure within the activity window and \nshould be corroborated with current network telemetry or additional intelligence sources when used for \ndefensive or investigative purposes.\n\nIP Addresses\n\nIndicator \n38.244.50[.]120\n\n144.172.64[.]60 \n144.172.64[.]99 \n144.172.64[.]128 \n187.15.88[.]186 \n187.15.90[.]217 \n187.15.89[.]186\n\nContext \nARTEX host and open directory  CrowdStrike⁶, Censys⁴ \n(South Korea financial \ncampaign) \nConnectWise campaign; sign-\nins and Graph API activity in \ncustomer environments \nGraph API access via \ncompromised apps (transient \nVPN IPs relevant October 15–\n16, 2026)\n\nTLP:CLEAR\n\nSource\n\nMicrosoft²\n\nMicrosoft²\n\n101.53.80[.]20 \n205.214.59[.]31 \n124.155.252[.]63 \n154.201.79[.]246 \n23.248.249[.]90 \n23.158.220[.]98 \n103.248.148[.]84 \n203.160.133[.]172 \n209.209.85[.]38 \n199.19.95[.]66 \n199.19.95[.]67 \n199.19.95[.]68 \n91.92.240[.]238 \n85.11.168[.]45 \n178.16.54[.]129 \n31.77.229[.]35 \n85.11.167[.]52\n\nDomains\n\nIndicator \nxcai[.]pro\n\nTLP:CLEAR\n\nProxy IP addresses (South  CrowdStrike⁶ \nKorea financial campaign)\n\nWISP C2 nodes (Storm-3358  Microsoft10 \ncampaign; active as of \nNovember 6, 2026) \nARTEX servers (Storm-3358  Microsoft10 \ncampaign; active as of \nNovember 6, 2026)\n\nC2 server (Storm-3358  Microsoft10 \ncampaign; active as of \nNovember 6, 2026)\n\nContext  Source \nLLM API reseller (South Korea  CrowdStrike⁶ \nfinancial campaign)\n\nNetwork and Behavioral Indicators\n\nIndicator Type  Description  Source \nOutbound API calls  api.anthropic.com,  Zscaler⁷ \napi.deepseek.com, \nopen.bigmodel.cn, api.x.ai \n(investigate unexpected calls \nalongside other suspicious \nactivity) \nHTTP request patterns  High-frequency automated  Zscaler⁷ \nrequests with sequential \nenumeration against \nlogin/search endpoints \nData transfers  Outbound transfers of  Zscaler⁷ \nstructured data (names, \nphone numbers, financial \ninformation) over HTTPS to\n\nTLP:CLEAR\n\nTLP:CLEAR\n\nunauthorized external \ndestinations \nWeb shells  *.aspx or *.jsp files in  Zscaler⁷ \nunexpected web application \npaths\n\nInformation Requested\n\nThe FBI encourages recipients of this document to report information concerning suspicious or criminal \nactivity to their local FBI field office at www.fbi.gov/contact-us/field-offices or the FBI Internet Crime \nComplaint Center (IC3) at www.ic3.gov.\n\nIf reporting an incident involving ARTEX or similar agentic SI frameworks, please provide the following \ninformation:\n\n1. ARTEX Infrastructure Artifacts: Source code or screenshots of suspected ARTEX console interfaces \nand ARTEX configuration files\n\n2. Authentication and Access Logs: Service principal sign-in logs showing unusual source networks, \nASNs, hosting providers, or geographies; user agent strings from suspicious authentication attempts; IP \nlogin information and associated timestamps; evidence of authentication from previously unseen \ninfrastructure\n\n3. Application and API Activity: Microsoft Graph activity; Microsoft Entra audit records; Microsoft 365 \naudit records showing reconnaissance; sequential lookups or unusually high numbers of records \naccessed per session on partner/employee portals; rate limit violations or volumetric anomalies in API \ncalls\n\n4. LLM and SI Infrastructure Indicators: Outbound connections to LLM API endpoints; evidence of self-\nhosted LLM gateway or reseller software (CLIProxyAPI, New API); unfamiliar LLM proxies or resellers in \negress logs; token consumption patterns or API billing anomalies\n\n5. Cloud and Container Activity: AWS IAM activity logs; Azure ARM deployment validation dry-runs across \nmultiple resource types; test blob write/delete operations in cloud storage with sub-second timing; \nevidence of compromised Kudu/SCM credentials or unusual Kudu API activity; container orchestration \nlogs; cloud metadata service queries; unexpected IAM role assumptions, PassRole operations, or instance \nprofile modifications\n\n6. Exploitation and Persistence Artifacts: Web shells; evidence of BLOODALCHEMY components; open-\nsource RAT indicators (WISP, vshell, AdaptixC2); newly created accounts with naming patterns;\n\nTLP:CLEAR\n\nunexpected EC2/Azure VM launches\n\nTLP:CLEAR\n\napplication credential modifications to OAuth applications; SSH key additions, Elastic IP allocations,\n\n7. Attack Characteristics: Compressed bursts of activity; high volumes of failed attempts with iterative, \nincremental changes; evidence of multiple sub-agents overwriting each other or generating duplicate \nrequests; mass notification/email broadcasts from compromised applications\n\n8. Loss and Impact Information: Total number of compromised accounts or affected users; types and \nvolumes of data accessed or exfiltrated; time from initial suspicious activity to detection; time from \ndetection to containment; total attempted loss amounts\n\nRecommendations\n\nThe FBI recommends organizations take the steps below to improve their security posture in response to \nARTEX and similar agentic SI frameworks.\n\nImmediate Actions\n\nReduce External Attack Surface:\n\n• Inventory all internet-facing applications, including peripheral systems: partner and supplier portals, \nmobile backends, employee applications, sales-support systems, legacy services, third-party integrations, \nand applications created across different business units.1,11\n\n• Restrict cloud administration, network device management, application publishing interfaces \n(Kudu/SCM endpoints), hypervisors, backup platforms, and other sensitive services to trusted networks, \nprivate connectivity, or dedicated administrative paths.⁵\n\n• Disable basic publishing authentication where not required; restrict access to deployment and \npublishing credentials; restrict SCM access to approved management networks or deployment \ninfrastructure.⁵\n\napplication files.⁵\n\nfeasible.7\n\nHarden Identity and Access:\n\n• Consider deployment models such as run-from-package that reduce direct modification of deployed\n\n• Establish governance around public IP allocation; require architecture review before workloads or \nmanagement interfaces are exposed directly to the internet.5\n\n• Move enterprise applications behind Zero Trust access solutions to eliminate external exposure where\n\n• Apply least privilege to multi-tenant application permissions and consent grants.2\n\nTLP:CLEAR\n\nTLP:CLEAR\n\n• Monitor credential additions, privilege assignments, and anomalous app-only activity, especially to \napplications with high privileged permissions and scopes.2\n\n• Review every Microsoft 365 tenant where affected ConnectWise applications are currently or were \npreviously installed or authorized, even if integration no longer in use; search by application ID rather than \ndisplay name.2\n\n• Review successful and failed service-principal sign-ins separately; investigate unfamiliar sources and \nactivity outside expected vendor infrastructure (do not limit review to published IPs).2\n\n• Review Microsoft Entra audit records for application/service principal credential additions; new \napplications/service principals/owners; app-role assignments and consent grants; privileged role \nassignments; changes to users/groups/authentication methods/Conditional Access (CA). If suspicious \nactivity is confirmed, revoke unnecessary permissions and rotate credentials.2\n\n• Require certificate manager or CA approval for higher-risk enrollment scenarios; enable Extended \nProtection for Authentication on CA web interfaces where supported; monitor unusual certificate \nenrollment involving privileged users or unexpected requesting systems.5\n\nStrengthen Application Protections:\n\n• Put guardrails around high-impact business APIs: rate limits, recipient limits, transaction thresholds, \nabuse controls appropriate to function; alert when application operates materially outside normal \nvolume.5\n\n• On partner and employee portals, alert on sequential lookups and unusually high numbers of records \naccessed per session; apply rate limits to affected endpoints; enforce server-side authorization for access \nto customer records and limit responses to fields required for the operation.7\n\n• Avoid returning detailed stack traces, compiler messages, SQL errors, internal paths, framework details \nto unauthenticated users; keep detailed troubleshooting information in internal logs.5\n\n• Review application configurations, environment variables, infrastructure-as-code, deployment scripts, \nCI/CD pipelines, databases for embedded secrets; implement continuous scanning of repositories, \ncontainer images, build artifacts, pipelines for exposed secrets; rotate exposed credentials promptly.5\n\nPreserve Evidence and Respond:\n\n• Preserve available Microsoft Entra sign-in and audit records, Microsoft Purview Audit records, SIEM data, \npreviously configured Microsoft Graph telemetry; do not delay urgent containment while waiting for \nexports.2\n\n• Review activity starting from at least October 1, 2026, for ConnectWise-related activity; extend service-\nprincipal sign-in review and investigate earlier related activity including failed authentications and volume \nanomalies.2\n\nTLP:CLEAR\n\nTLP:CLEAR\n\n• If suspicious activity is confirmed, have appropriately authorized responder remove confirmed \nunauthorized credentials, grants, owners, assignments, identities after preserving evidence.2\n\n• Keep affected ConnectWise application connections disabled until ConnectWise provides formal \nrestoration instructions; do not delete enterprise applications as first response (preserve configuration \nand evidence before irreversible changes).2\n\nTLP:CLEAR\n\nTLP:CLEAR\n\nNotes\n\n_______________________\n\n1. Akamai, \"What ARTEX Reveals About Overlooked Attack Surfaces,\" October 23, 2026.\n\n2. Microsoft Security Research and Microsoft Threat Intelligence, \"AI-driven, large-scale nation-state \ncampaign using ARTEX enables supply chain compromise,\" October 21, 2026.\n\n3. Google Threat Intelligence Group, \"ARTEX and Friends: Enter Agentic Offensive Frameworks,\" October \n30, 2026.\n\n4. Censys, \"ARTEX: Agentic Asbestos,\" November 5, 2026.\n\n5. Google Threat Intelligence Group and Mandiant, \"ARTEX and Friends: Enter Agentic Offensive \nFrameworks – Defensive Strategies from Mandiant Consulting,\" October 30, 2026.\n\n6. CrowdStrike, \"Unknown Threat Actor Uses AI-Driven ARTEX to Target South Korean Finance,\" October 7, \n2026.\n\n7. Zscaler ThreatLabz, \"Threat Actor Abuses Agentic AI Pentesting Framework to Target South Korean \nFinancial Institutions,\" October 14, 2026.\n\n8. CERT-EU, \"From Penetration Testing to Cyber Threat: Beware of ARTEX,\" November 5, 2026.\n\n9. Check Point Research, \"Hacking on Autopilot: Inside ARTEX Cybercrime Campaigns,\" November 5, \n2026.\n\n10. Microsoft Threat Intelligence, \"ARTEX abuse expands as cybercriminals target financial sector,\" \nNovember 2026.\n\n11. Akamai, \"What ARTEX Reveals About Overlooked Attack Surfaces – API Security Implications,\" October \n23, 2026.\n\nTLP:CLEAR\n\nTLP:CLEAR\n\nReporting Notice\n\nIf you identify suspicious activity within your enterprise or have information related to the contents of this \ndocument, please contact your local FBI Cyber Squad immediately at www.fbi.gov/contact-us/field-\noffices. The FBI also encourages you to report suspicious or criminal activity to the FBI Internet Crime \nComplaint Center at www.ic3.gov. When available, each report should include the date, time, location, \ntype of activity, number of people, and type of equipment used for the activity, the name of the submitting \ncompany or organization, and a designated point of contact. Press inquiries should be directed to the FBI’s \nNational Press Office at npo@fbi.gov or (202) 324-3691.\n\nIndividual indicators included in this document should always be evaluated in light of your complete \ninformation security situation. Some indicators, particularly those of a nondeterministic or ephemeral \nnature (such as filenames or IP addresses), may not be indicative of a compromise.\n\nYour organization has no obligation to provide information in response to this product. If, after reviewing \nthe information provided, your organization decides to provide information to the FBI, it must do so \nconsistent with applicable state and federal law.\n\nAdministrative Note\n\nThe information in this document is being provided by the FBI, with no guarantees or warranties, for \npotential use at the sole discretion of recipients to protect against cyber threats. This data is provided to \nhelp cybersecurity professionals and system administrators guard against the persistent malicious actions \nof cyber actors. The FBI does not endorse any commercial entity, product, company, or service, including \nany entities, products, or services linked within this document. Any reference to specific commercial \nentities, products, processes, or services by service mark, trademark, manufacturer, or otherwise, does \nnot constitute or imply endorsement, recommendation, or favoring by the FBI.\n\nThis FLASH is marked TLP:CLEAR. The information in this product may be shared without restriction. \nInformation is subject to standard copyright rules.\n\nYour feedback regarding this product is critical. \n \nPlease take a moment to complete the survey at the link below. Input can be submitted\n\nanonymously and should be specific to your experience with our written products.\n\nhttps://www.ic3.gov/PIFSurvey\n\nThis survey is for feedback on contact and value only. Reporting of technical information\n\nregarding FLASH reports must be submitted through your local FBI field office.\n\nTLP:CLEAR",
	"extraction_quality": 1,
	"language": "EN",
	"sources": [
		"OTX"
	],
	"origins": [
		"pdf"
	],
	"references": [
		"https://www.ic3.gov/CSA/2026/261113.pdf"
	],
	"report_names": [
		"261113.pdf"
	],
	"threat_actors": [],
	"ts_created_at": 1794881499,
	"ts_updated_at": 1794881985,
	"ts_creation_date": 1794574626,
	"ts_modification_date": 1794574626,
	"files": {
		"pdf": "https://archive.orkl.eu/06413330a7f7a014d4da313667428546df094103.pdf",
		"text": "https://archive.orkl.eu/06413330a7f7a014d4da313667428546df094103.txt",
		"img": "https://archive.orkl.eu/06413330a7f7a014d4da313667428546df094103.jpg"
	}
}