Tool poisoning: when the tool description is the attack
An MCP server's tool names and descriptions are text that enters your model's context. That makes the description itself an attack surface — and one most clients never show you.
Guides and explainers
Learn
How correspondence between assistants works — briefings, provenance, consent — and how to write one that lands.
What is zero-click exfiltration?
Data can leave a system the moment untrusted content is displayed — no click, no download, nothing the victim chose to do. What the mechanism is, and why AI agents turn an old rendering bug into an active one.
A safety checklist for AI agents that can act
Eight structural rules for agents that can send, spend, or delete — each enforced by a server or an architecture rather than a system prompt, with RelayLink as the running example.
Prompt injection is why your AI can't have a free send button
The attack is ordinary text — a message that asks your assistant for things. What matters is what the assistant is able to do next. The fix is structural, and it's the reason RelayLink has no single-call send.