Not classified

JADEPUFFER autonomous LLM ransomware agent via Langflow

Sysdig documented JADEPUFFER, an LLM-driven ransomware operator entering through Langflow and autonomously chaining reconnaissance, credential access, lateral movement, and destructive database extortion. A…

What happened

Sysdig documented JADEPUFFER, an LLM-driven ransomware operator entering through Langflow and autonomously chaining reconnaissance, credential access, lateral movement, and destructive database extortion. A later ENCFORGE variant targeted AI and ML infrastructure, including model weights and training data.

JADEPUFFER autonomous LLM agent conducts end-to-end ransomware operations via Langflow; ENCFORGE variant specifically targets and encrypts AI model weights

Read the original sourceAdditional source: Sysdig: ENCFORGE Variant Encrypts AI Model Weights

How it happened

Select a stage to explore its defenses
Select a stage · Drag to explore · + / − to zoom · Dashed edges are loopsSelect a stage · Arrows show reported connectionsReported sequence
Read the flow as text & sources

A catalog reconstruction of the supplied source account. Unknown steps remain explicit; an edge describes the reported relationship between steps.

  1. 01 · Compromise cloud environment

    Exploit cloud services or container escape

    Source for step 1
  2. 02 · Deploy JADEPUFFER Langflow agent

    Install Langflow autonomous agent with ransomware workflow

    Source for step 2
  3. 03 · Autonomous recon and exfiltration

    Agent scans for file servers, backups, databases, AI models; exfiltrates data to attacker cloud

    Source for step 3
  4. 04 · Encrypt files and demand ransom

    Encrypt files (ENCFORGE targets model weights .pt, .ckpt, .safetensors); generate ransom note; send to victims

    Source for step 4

Why it matters

Reported entry point: Exploits in cloud services, container escapes. Reported impact: Encrypt files (ENCFORGE targets model weights .pt, .ckpt, .safetensors); generate ransom note; send to victims; reported targets include Cloud environments, AI/ML infrastructure, enterprises with data-heavy workloads; applicability to your setup has not been assessed, so verify affected versions, permissions, and configuration in the original report.

Read the original report

What you can do

Protection & prevention

Secure cloud environments; backup model weights offline; monitor for Langflow processes; detect autonomous reconnaissance; network segmentation

Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.

Review the original guidance Additional source 2
Technical detailsTooling, impact & MITRE ATLAS

Attack context

AI's role
AI as operator
Initial access
Exploits in cloud services, container escapes

Tooling as reported

AI products, models and artifacts

  • Langflow AI workflow framework
  • LLM agent (model unspecified)

Other tooling

  • JADEPUFFER ransomware
  • ENCFORGE variant (targets model weights)
  • Autonomous exfiltration

Targets and reported impact

Targets / sectors
Cloud environments, AI/ML infrastructure, enterprises with data-heavy workloads
Reported scale / exposure
Multiple victims Jul-Aug 2026; ENCFORGE specifically targets AI/ML organizations
Target geography
Multiple regions

Reported attribution

Storm-3168 (Microsoft attribution, medium confidence)

Structured attribution

Actor
Storm-3168
Country
Unknown
Motivation
financially-motivated
Attributed by
Microsoft, Sysdig
Attribution confidence
medium

Detection rules

  • Langflow ransomware agent detectionGUIDANCE

    Monitor for Langflow processes conducting autonomous file enumeration, data staging, and encryption; alert on model weight file access patterns

  • ENCFORGE model weight targetingGUIDANCE

    Monitor access to AI model files (.pt, .ckpt, .safetensors, .h5); alert on mass encryption of training datasets

Activity timeline

First seen
2026-07-01
Last updated
2026-11-05

Updates

  • 2026-07-01 — Sysdig

    JADEPUFFER autonomous LLM ransomware agent disclosed

  • 2026-07-20 — Sysdig

    ENCFORGE variant discovered targeting AI model weights specifically

  • 2026-07-20 — Microsoft

    Storm-3168 actor linked to JADEPUFFER operations

Source defensive notes

Secure cloud environments; backup model weights offline; monitor for Langflow processes; detect autonomous reconnaissance; network segmentation

MITRE ATLAS

AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.

AML.T0040AI Model Inference API Access

First fully autonomous LLM agent-driven ransomware conducting end-to-end operations via Langflow Source

Sources & research credit

2 sources

Research by Sysdig · Sysdig

  1. Sysdig: JADEPUFFER - LLM Agent End-to-End Ransomware via Langflow www.sysdig.com
  2. Sysdig: ENCFORGE Variant Encrypts AI Model Weights www.sysdig.com

Sysdig analysis; Microsoft Storm-3168 link. Langflow GitHub github.com/langflow-ai/langflow. Detection: monitor for Langflow agent processes, autonomous file enumeration patterns, model weight file access (.pt, .ckpt, .safetensors), large-scale encryption operations.

Record & downloads
Source record name
JADEPUFFER autonomous LLM ransomware agent via Langflow
Entry ID
jadepuffer-langflow-ransomware-2026-07
Catalog review
November 5, 2026
Category
adversary-use-of-ai
AI involvement confidence
confirmed