Hackers target LiteLLM and Starlette vulnerabilities for unauthenticated code execution
SANS Internet Storm Center detected active scans targeting CVE-2026-42271 and CVE-2026-48710 vulnerabilities to achieve unauthenticated remote code execution.
FTMQ Security, written by our newsroom0 views

SANS Internet Storm Center reported that automated scans began targeting honeypots over the weekend of November 14 to November 15, 2026. The requests targeted the path /mcp-rest/test/tools/list to exploit two security flaws, CVE-2026-42271 and CVE-2026-48710. Cyber research firm Horizon3.ai published an analysis showing that hackers can chain these vulnerabilities together to achieve remote code execution without authentication. [1]
LiteLLM is an open-source large language model gateway developed by BerriAI, which was founded by Ishaan Jaffer and Krrish Dholakia. The software acts as a proxy server or Python library to provide a unified application programming interface for routing requests across multiple AI providers. [2]
LiteLLM relies on underlying web infrastructure including FastAPI, a Python web framework introduced in 2018 for building application programming interfaces. According to metrics on its official website, LiteLLM has recorded over 240 million Docker pulls and processed over one billion requests. Enterprise adopters of the tool include Netflix and Lemonade. [2][3][4]
Security administrators managing LiteLLM proxy deployments should inspect network logs for unauthorized HTTP requests containing the /mcp-rest/test/tools/list endpoint path. System teams are advised to review exposed instances and apply security updates for the affected LiteLLM and Starlette components immediately to prevent unauthenticated server compromises. [1][2]
In short
- SANS Internet Storm Center detected active scans targeting CVE-2026-42271 and CVE-2026-48710 in LiteLLM and Starlette.
- Horizon3.ai documented that chaining the two vulnerabilities allows unauthenticated remote code execution.
- The malicious requests observed in honeypots target the path /mcp-rest/test/tools/list.
- LiteLLM is an open-source AI gateway developed by BerriAI with over 240 million Docker pulls.
Sources
Every paragraph above points to the numbered items it rests on. Read the originals here.
- [1]Scans for LiteLLM/Starlette (CVE-2026-42271, CVE-2026-48710), (Mon, Nov 16th)SANS Internet Storm Center, 1d ago (the report this story comes from)
Background
Our newsroom writes these reports with the help of software, from the 4 sources listed and nothing else, and checks them against those sources. Facts can still be wrong or move on; the originals are the record. Spotted a mistake? Write to daniel@monsterkong.com.
Related from FTMQ Security
Earlier reports of ours on the same people and subjects.
- Hackers target GPUStack authentication bypass vulnerability CVE-2026-102895 in AI clustersThreat Intelligence, 5d ago
- SANS Warns of Exploitation Involving Legacy CGI Email Vulnerability CVE 2000 0411Threat Intelligence, 13d ago
- Microsoft releases fix for high severity Paint code execution vulnerabilityBugs and Patches, 9d ago
- Storm 2603 deployed ransomware through Microsoft SharePoint vulnerabilitiesNation State Hacking, 9d ago
More in Threat Intelligence
- CISA warns of security flaws in Dorsett Controls InfoScan client4h ago
- Siemens patches code injection vulnerability in ROS software library4h ago
- Industrial ransomware incidents increased 11 percent in third quarter of 20261d ago
- Cyberespionage group targets Southeast Asian government entities using HTTP3 RATs1d ago
- Chinese speaking hackers use AI framework Cairn in 25 attack operations2d ago
- WardenStealer Infostealer Targets Local AI Session Credentials and API Keys2d ago
Get the day in one email
Reports like this one, the top news of the last 24 hours, every morning. Free, one email a day; readers can comment under every report.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.

Comments
Loading
Join the conversation
Comments are open to readers of our daily email: the top news of the last 24 hours, every morning, free. Sign up and the comment box opens.
Already on the list? Enter the same address and we will send a sign-in link.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.