AI Risk Wire

Cases › security breach › LR-2026-0883

AntiFox agentic framework intrusions against AWS credentials (as reported via GitGuardian blog)

Who did it
Outside attacker
Who was hit
internet-facing web applications and AWS environments
AI agent used
AntiFox
How it was found
Third party
Reported
11 Nov 2026
Severity
Major
  1. Attackers ran the AntiFox multi-agent offensive framework against web applications and AWS environments, using valid IAM user and assumed EC2 role credentials to enumerate cloud services and test permissions.
  2. Attackers used the framework to compromise web applications, steal configuration data and use valid AWS identities, generating hundreds of API requests in about 30 minutes.
  3. Some requests succeeded and exposed sensitive AWS resources, including production secrets.
  4. Arctic Wolf researchers investigated and linked infrastructure hosting AntiFox to attacks on web applications and AWS environments.

Source: GitGuardian Blog · 7 more

Risks Credential and privilege abuse (35 incidents) · AI system attacked or breached (54 incidents)

No source states: AI company and model, money lost or records, tools it had, company using the AI, sector.

Evidence

Sources (8)

  1. AntiFox: ARTEX AI Fork Targets AWS Credentials · GitGuardian Blog · 2026-11-11
  2. Attackers use AntiFox, an open-source AI pentest agent run from Alibaba Cloud, to steal AWS secrets, Arctic Wolf says · MLQ.ai · 2026-11-11
  3. Hackers use AntiFox AI agents in AWS credential-theft attacks · BleepingComputer · 2026-11-11
  4. Hackers Use AntiFox AI Hacking Tool to Compromise AWS Cloud Environments and Steal Sensitive Data · Cyber Press · 2026-11-12
  5. Hackers Abuse AntiFox AI Pentesting Tool to Deploy Webshells and Access AWS Production Secrets · CyberSecurityNews · 2026-11-12
  6. AntiFox AI Agents Linked to AWS Attacks That Exposed Production Secrets and Compromised Web Servers · GBHackers News · 2026-11-12
  7. Attackers are Using Agentic AI Offensive Tools to Automate Intrusions · Security Boulevard · 2026-11-12
  8. Ex-FBI Agent Steals $2.5M in Crypto, Grok App Data Exposures and More Cybersecurity News | ForkLog · ForkLog · 2026-11-14
Show every field, with its source and grade

The AI system

Software around the model
AntiFox
C one secondary source states it
What it was allowed to do
Use any valid AWS identities the framework could find, after stealing configuration data.
C one secondary source states it
What it was for
offensive agentic AI framework; self-described AI autonomous penetration testing platform
D inferred by us from the described behaviour, reason shown

What happened

Earliest source we hold
2026-11-11
D inferred by us from the described behaviour, reason shown
Who caught it
an outside party
D inferred by us from the described behaviour, reason shown
Themes
agentic attack framework, credential theft, AWS cloud access
C one secondary source states it
Who caused it
an outside attacker
D inferred by us from the described behaviour, reason shown
What went wrong
security breach
D inferred by us from the described behaviour, reason shown
AI agent
an AI agent acted
C one secondary source states it

Impact

Severity
major
D inferred by us from the described behaviour, reason shown
Kind of harm
others' security
D inferred by us from the described behaviour, reason shown
Who was affected
internet-facing web applications and AWS environments
C one secondary source states it
Kinds of data exposed
credentials
D inferred by us from the described behaviour, reason shown
Why this severity
Some attacker requests succeeded, including operations that exposed production secrets, a live production compromise.
D inferred by us from the described behaviour, reason shown

LR-2026-0883 · version 13 · published 2026-11-15 after automatic checks; not yet checked field by field by a person