Skip to main content
Live Webinar with SANS (June 25)— Agentic CTI Automation for Fun & ProfitRegister Free
Mallory

Chaos Agent

Also known asChaos Agent

Chaos Agent is the name used by Pillar Security for the hackerbot-claw campaign that operated from February 21 to February 28, 2026 and targeted developer environments via GitHub Actions. The campaign exploited the GitHub Actions pull_request_target permission model by forking repositories, embedding payloads in a branch name or CI script, and submitting trivial pull requests to trigger workflows and harvest secrets. Reported targets included repositories belonging to Microsoft, Datadog, and Aqua Security. In the Aqua Security case, StepSecurity confirmed theft of a Personal Access Token and subsequent repository takeover. The activity also led to malicious code being injected into Trivy VS Code extension versions 1.8.12 and 1.8.13 distributed through Open VSX. This compromise, identified as CVE-2026-28353, was described as the first documented weaponization of locally installed AI coding CLIs against developer environments. Abused tools included Claude, Codex, Gemini, GitHub Copilot CLI, and Kiro. Version 1.8.12 used compromised AI coding CLIs to collect system data, while version 1.8.13 improved exfiltration by routing data to a GitHub repository named posture-report-trivy using the victim's authenticated GitHub CLI session. The campaign focused on developer trust zones rather than production networks, seeking access to high-value credentials and secrets commonly present in developer machines and CI/CD systems. No additional aliases or sub-groups beyond hackerbot-claw and Chaos Agent are directly supported by the provided content.

Share:
Are they targeting you?

Know when an actor pivots toward your sector

Mallory correlates actor tradecraft and target patterns against your stack, your sector, and your geography. See overlap before they land.

OPERATIONAL PROFILE

Targeting

Who, where, and (when attributed) which flag flies behind the operation. Pulled from open-source reporting and Mallory's analyst review.

Who they target

Sectors the actor has been observed targeting.

  • Software & Services
MITRE ATT&CK

Tradecraft

5 distinct techniques observed across reporting, grouped by tactic. Hover any cell for the evidence excerpt; click through for MITRE's full description.

4 of 15 tactics5 techniques×N= number of intelligence reports citing this technique
MITRE ATT&CK
TA0001
Initial Access
1 technique
T1195
Supply Chain Compromise
TA0002
Execution
1 technique
T1059
Command and Scripting Interpreter
TA0006
Credential Access
2 techniques
T1212
Exploitation for Credential Access
T1649
Steal or Forge Authentication Certificates
TA0010
Exfiltration
1 technique
T1567
Exfiltration Over Web Service
What this page doesn’t show

The version that knows your environment.

This page is what’s public. Mallory adds the parts that aren’t: sector and geo overlap with your footprint, the IOCs they’re burning right now, detection coverage, and what to do next.
Target overlap

Match sector + geo + tech-stack targeting against your real footprint.

Tradecraft mapping5

Every observed MITRE ATT&CK technique, grouped by tactic.

Malware arsenal

Families this actor is known to deploy, with IOCs and behavior.

Exploited CVEs

CVEs this actor has used in known campaigns.

Detection signatures

YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.

Observables

Domains, IPs, and hashes tied to this actor, refreshed continuously.