This week in the Wild West Roundup:

2026-08-31 · Bits and Bobs 8/31/26

This week in the Wild West Roundup:
- Breaking Claude Code Opus 5 Auto Mode.
  - "A simple website summary request hijacks Claude Code Opus 5 in Auto Mode and achieves code execution with 60-80% attack success rate using a small sample size."
- Paper: InjecMEM: Memory Injection Attack on LLM Agent
  - It allows hackers to plant hidden instructions in AI memory with a single prompt.
- A Malicious Webpage Could Poison Your Local AI Model Behind NVIDIA NemoClaw.
- Power Leak: Amazon Kiro IDE Prompt Injection Enables Data Exfiltration.
- Attackers Exploit MCP RCE, Blind Prompt Injection and Memory Credential Theft Against AI Infrastructure.
- Prompt injection ranks No. 1 with OWASP and No. 12 in the incident record.
  - But part of that is that the attack is invisible to a scan, structurally!
- An agent changed its company's production DNS settings.
  - Whoops!

More on this topic

From other episodes