6
min read
Research

Memory Poisoning & Instruction Drift: From Discord Chat to Reverse Shell (OpenClaw Hackathon Findings)

A controlled experiment showing how persistent memory and instruction drift in an AI agent led to reverse shell execution via Discord alone.
Platon Frolov
February 18, 2026
Activate
untouchable mode.
Get started for free.

Lakera Guard protects your LLM applications from cybersecurity risks with a single line of code. Get started in minutes. Become stronger every day.

Join our Slack Community.

Several people are typing about AI/ML security. 
Come join us and 1000+ others in a chat that’s thoroughly SFW.