Three Distinct Exploitation Points, Not One Generic Threat
These three attack classes are often discussed as if they were variations of the same problem. They are not. Each targets a different component of an AI agent's architecture, and each requires its own detection logic and its own runtime controls rather than a single generic defense.
Prompt injection manipulates the input context an LLM reasons over, either directly through user-supplied text or indirectly through retrieved external content merged into the prompt. Tool poisoning exploits trust in tool or MCP metadata that the model uses to select and invoke actions, before any execution decision is made. Memory poisoning corrupts persistent memory stores so that the effects of a compromise surface later, in sessions unrelated to the original attack.
Attack Surface Summary
| Attack type | Primary surface | Mechanism |
|---|---|---|
| Prompt injection | Input/context-assembly layer | Targets where user input and retrieved content merge into the model prompt, affecting the active context window. |
| Tool poisoning | Tool/MCP registration and discovery layer | Exploits trust in tool or function-calling metadata ingested by the model before any execution decision. |
| Memory poisoning | Persistent memory read/write boundary | Corrupts |