Darktrace researchers turned Claude Code into a remote-controlled implant in a lab test, steering the coding agent through the memory stores feature of Anthropic’s Claude platform and pulling a 1 GB file out through the Claude Files API. Once the agent was set up, every job and every result traveled through api.anthropic.com, the endpoint the agent already calls for inference, so the attacker’s traffic looked like a coding agent doing its job. The technique needs … More
The post Stolen Claude API keys can turn Claude Code into a remote-controlled implant appeared first on Help Net Security.
source & further reading
helpnetsecurity.com — original article
AWS Emulator: Open-source tool for testing AWS apps locally, built by autonomous AI agents
Bots pose as AI agents, AI agents pose as humans, and defenses believe both
Kai extends Autonomous Defense Platform visibility to internet-facing assets