ARTEX: Attackers Adopted This AI Pentesting Tool in Three Months. Leaked Secrets Are Its Fastest Way In.

Summary

ARTEX, an open-source tool where AI agents carry out penetration tests with minimal human steering, has been adopted by state-aligned groups, criminal crews and a runaway research agent since August 2026. Reporting cites hundreds of targeted organizations, and one agent moved from a GitHub search for leaked secrets to stolen cloud data in five minutes. Attackers most often gain entry through exposed credentials, including a guessable development password that reached production and a backdoored ARTEX copy whose Docker image was downloaded more than 9,000 times.

Mitigation

Scan public and internal repositories for secrets, including git history, and rotate every secret found. Replace long-lived cloud keys in CI secrets with short-lived OIDC sign-in tokens, set expiry on personal access tokens, and ensure no internet-facing system accepts a development or default credential. Identify any ARTEX copy and Docker image installed in your organization, and hunt for the listed indicators, starting with the user agents artex-enrich/1.0 and norma/0.4.

Related items

• MediumARTEX and beyond: AI-enabled penetration testing platformsSimilar attack · Cisco Talos Blog
• MediumAmazon Threat Intelligence observes threat actors running cybercrime campaigns with ARTEX AI agentsSimilar attack · AWS Security Blog
• InfoHow AI is changing the nature of cyberattacksSimilar attack · CSO Online
• MediumInside the loop: Five AI attack frameworks that never ask permissionSimilar attack · Elastic Security Labs
• MediumHow one AI agent exposed five critical enterprise security gapsSimilar attack · BleepingComputer