Definition
Plain language
A security research firm that documents real-world attacks against AI assistants.
As stated in the literature
An AI-security research group that published the 2025 Agent Flayer attack chain, among other indirect-prompt-injection findings against MCP-connected agents.
Why it matters: Their published attacks raise the real-world urgency of treating agent ecosystems as security-critical, not just research playgrounds.
For example, Zenity Labs demonstrated the 2025 Agent Flayer chain showing how an MCP-connected assistant could be hijacked via crafted content.
Heard on the show
“That attack chain has a name in the wild — it's called Agent Flayer, documented by Zenity Labs in twenty-twenty-five — and the paper we're talking about today emulates it as one of its test cases.”Episode 057 — How Uber Caught 206 Leaked Credentials With an LLM-Powered Security Stack