Skip to main content
Mallory
← Back to News
ai-enabled-threat-activitylateral-movement-methodinitial-access-methodpersistence-method

Cloudflare Reports Surge in AI Pentesting Tools and ARTEX Abuse of Stolen API Tokens

Emerging Technology Risk · 2 sources · Updated 2d ago · First seen Nov 13, 2026

Cloudflare Reports Surge in AI Pentesting Tools and ARTEX Abuse of Stolen API Tokens

Cloudflare reported observing more than 50 families of autonomous, LLM-driven penetration-testing and exploitation frameworks scanning its customers from thousands of IP addresses. Daily requests from observed tools increased more than thirtyfold since October 1, and ARTEX and its forks represented more than 60% of known AI penetration-testing orchestrators as of November 12. The findings highlight the growing use of AI-orchestrated tools to probe and attack exposed infrastructure.

Cloudforce One investigated incidents in which attackers used compromised, static CI/CD credentials to move from hyperscaler environments into customers’ Cloudflare environments. In one investigation, ARTEX used a compromised account-level API token to create Worker scripts, issue database dump requests, attempt persistence, and validate session tokens for third-party applications. Cloudflare recommended layered defenses combining identity-gated access, micro-segmentation, traffic filtering, API validation, and threat intelligence, underscoring the importance of protecting automation credentials alongside internet-facing systems.

New surface, new adversariesTrack how attackers are adapting to this technology.
Track this risk
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

5 events · Oct 8, 2026 – Nov 12, 2026
  1. Reported Nov 13, 20264d ago

    Cloudflare detects widespread scanning by autonomous AI security frameworks

    Cloudflare observed more than 50 LLM penetration-testing and exploitation framework families scanning customers globally over the preceding three days, operating from thousands of IP addresses. Its analysis found more than 12,000 repositories containing ARTEX lineage and a greater than thirtyfold increase in daily tool requests since October 1.

    Adapting to AI penetration testing orchestrators: The Modern Security Model | Cloudflare Blog
  2. Reported Nov 13, 20264d ago

    ARTEX uses compromised Cloudflare API token for exploitation and persistence attempts

    In one investigated incident, ARTEX used a compromised account-level CI/CD API token without IP restrictions or rate limiting to create Cloudflare Worker scripts and issue 400 database dump requests within one hour. Investigators also observed attempted persistence and validation of session tokens for third-party applications, assessing that the activity was lightly guided by a human operator.

    Adapting to AI penetration testing orchestrators: The Modern Security Model | Cloudflare Blog
  3. Reported Nov 13, 20264d ago

    Cloudforce One traces cross-cloud intrusions to compromised CI/CD credentials

    During multiple incident-response investigations over the preceding eight weeks, Cloudforce One observed attackers pivoting from hyperscaler environments into customers’ Cloudflare environments. Root-cause analysis identified compromised static, non-expiring CI/CD tokens with broad infrastructure access.

    Adapting to AI penetration testing orchestrators: The Modern Security Model | Cloudflare Blog
  4. Nov 12, 20266d ago

    ARTEX and its forks exceed 60% of known AI penetration-testing orchestrators

    Cloudflare found that ARTEX and its forks accounted for more than 60% of currently known AI penetration-testing orchestrators as of November 12.

    Adapting to AI penetration testing orchestrators: The Modern Security Model | Cloudflare Blog
  5. Oct 8, 20261mo ago

    ARTEX creator removes original repository over malicious use

    ARTEX creator Autumn-27 removed the original repository, stating that malicious exploitation contradicted the tool’s original intent.

    Adapting to AI penetration testing orchestrators: The Modern Security Model | Cloudflare Blog
Do you run this?Follow how adversaries are adapting to this technology, and where it touches your stack today.
Track This Risk
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.