A security analysis revealed a ClawHub skill, clawrtc, that installs a crypto miner on user machines. While the skill's disclosure omits that it sends hostnames and MAC addresses, it does request consent for data transmission. The miner can bypass DNS-based security by connecting directly via IP address, and a flag allows it to start automatically on login. Researchers recommend monitoring agent session activity and network connections to detect such malicious skills. AI
IMPACT Highlights the need for robust security practices and runtime monitoring for AI agent skills to prevent the installation of malicious software.
RANK_REASON Analysis of a specific malicious skill within an AI agent ecosystem.
- bottube.ai
- ClawHub
- clawrtc
- Manifold Security
- Python Package Index
- rustchain.org
- Swapnil Patil
- ThreatFox
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →