Executive Summary
Oasis Security disclosed a weakness in NVIDIA NemoClaw that lets an attacker-controlled webpage take unauthenticated control of the local Ollama instance serving an AI agent and plant hidden instructions in the model.
What Happened
Oasis Security has disclosed a weakness in NVIDIA NemoClaw that could let an attacker-controlled webpage take unauthenticated control of the local Ollama instance serving an AI agent and plant hidden instructions inside the model itself. The findings were shared with The Hacker News ahead of publication, and the report says Oasis Security reported them to NVIDIA's Product Security Incident
Affected Products
- NVIDIA NemoClaw
- Ollama
Vulnerability / Attack Technique
Unauthenticated control of a local model server from a webpage
Technical Mechanism
A webpage in the user's browser can reach the local Ollama API without authentication and modify the model's instructions.
AI Security Category
LLM04:2025 Data and Model Poisoning
Part of the AI Supply Chain and AI SBOM topic.
Relevant Frameworks
RealGround Analysis
Local does not mean private when a browser can reach it. RealGround analysis: authenticate local model APIs and block cross-origin access.
Recommended Controls
- Require authentication on local Ollama instances
- Block browser access to local model ports
- Apply NVIDIA's guidance when published
Related RealGround Service
Secure AI Agent Build · or start with the free AI Agent Security Scanner.
Sources
thehackernews.com: https://thehackernews.com/2026/08/a-malicious-webpage-could-poison-your.html
Summary and analysis written by RealGround from the source above; last updated 2026-10-17.