---
title: Amazon: a suspected Chinese actor told Cairn AI agents to 'obtain shell access', and they found and exploited a zero-day in MONAI Label to get root | ctxsec
url: https://ctxsec.com/items/amazon-threat-intel-chinese-nexus-cairn-ai-agents-found-monai-label-zero-day-root-shell-east-asian-universities
description: "Amazon Threat Intelligence found an unauthenticated, internet-exposed instance of the open-source Cairn autonomous pentesting framework on September 10 and read its history. Over about 48 hours from September 8, the operator pointed agents at East Asian university IPs with goals like \"obtain shell access.\" Against an open-source medical imaging AI app (MONAI Label), the agents mapped the API and exploited an unreported path traversal in an upload endpoint within about 11 minutes, then wrote a reverse shell into sitecustomize.py and triggered a training job to get a root shell. Other runs failed: 16 hours and 148 actions against an open Redis server, and more than 30,000 credential attempts over six hours against a GPU cluster. Amazon assesses with moderate confidence the actor is China-nexus, based on Simplified Chinese prompts, CST timestamps and targeting."
status: "official (Amazon Threat Intelligence, AWS Security Blog, Nov 20, 2026; MONAI Label GHSA-qc22-rmgc-vg94); attribution is Amazon's moderate-confidence assessment, last updated Nov 20, 2026, 11:40 PM ET"
published: 2026-11-20
updated: 2026-11-20T23:40:00-05:00
---

> Amazon Threat Intelligence found an unauthenticated, internet-exposed instance of the open-source Cairn autonomous pentesting framework on September 10 and read its history. Over about 48 hours from September 8, the operator pointed agents at East Asian university IPs with goals like "obtain shell access." Against an open-source medical imaging AI app (MONAI Label), the agents mapped the API and exploited an unreported path traversal in an upload endpoint within about 11 minutes, then wrote a reverse shell into sitecustomize.py and triggered a training job to get a root shell. Other runs failed: 16 hours and 148 actions against an open Redis server, and more than 30,000 credential attempts over six hours against a GPU cluster. Amazon assesses with moderate confidence the actor is China-nexus, based on Simplified Chinese prompts, CST timestamps and targeting.

[← AI](/ai) · [markdown](/items/amazon-threat-intel-chinese-nexus-cairn-ai-agents-found-monai-label-zero-day-root-shell-east-asian-universities.md)

official (Amazon Threat Intelligence, AWS Security Blog, Nov 20, 2026; MONAI Label GHSA-qc22-rmgc-vg94); attribution is Amazon's moderate-confidence assessment

2026-11-20 3:49 PM ET · updated

Nov 20, 2026, 11:40 PM ET

### Amazon: a suspected Chinese actor told Cairn AI agents to 'obtain shell access', and they found and exploited a zero-day in MONAI Label to get root

- **The bug (now CVE-2026-133968, assigned Nov 20):** MONAI Label ≤0.8.5 lets `PUT /datastore/image` and `/datastore/label` write files anywhere the process can; auth is off by default. The fix is merged (PR #1915) but, per the advisory, ships in the next release. See the tip.
- **Defender takeaway (Amazon):** agents keep going around the clock, so put high-confidence detections on automated response.
- **Context:** Cairn abuse was earlier reported by Gambit Security (retail card theft), Unit 42 and Google.
- IOCs and hunting: see the hunt item.

Sources: [AWS Security Blog (CJ Moses): Chinese-nexus actor's Cairn AI agents found a zero-day on their own (Nov 20, 2026)](https://aws.amazon.com/blogs/security/chinese-nexus-actors-cairn-ai-agents-found-a-zero-day-on-their-own/) · [MONAI Label GHSA-qc22-rmgc-vg94: Path traversal in datastore upload endpoints allows arbitrary file write (critical, Nov 18, 2026)](https://github.com/Project-MONAI/MONAILabel/security/advisories/GHSA-qc22-rmgc-vg94) · [CVE-2026-133968 record (MITRE, published Nov 20, 2026)](https://www.cve.org/CVERecord?id=CVE-2026-133968)
