Skip to content

Hide Navigation Hide TOC

MCP Line Jumping — Agent-Directed Imperative Embedded in a Tool/Parameter Description Field (Pre-Invocation Injection) - ATR-2026-00579 (164e9fea-5134-5340-8eb6-b318a79b3f2c)

Detects the MCP "line jumping" attack class (The Vulnerable MCP Project entry line-jumping-attack, reported by Trail of Bits). A malicious MCP server smuggles instructions aimed at the model INTO A TOOL-SCHEMA OR PARAMETER DESCRIPTION FIELD. Because MCP clients load every tool description into the model's context the moment a server is listed, the injected instruction executes BEFORE the tool is ever invoked — jumping the line ahead of user approval of any tool call. The detectable signature is a tool/parameter schema "description" field whose value carries an agent-addressed pre-invocation imperative: telling the assistant/model what it MUST do (prepend a command, route output, ignore the user) before or whenever it calls a tool. This is distinct from a conversation-level "ignore previous instructions" (the directive must live inside a tool-schema description field) and from the rug-pull class (no temporal redefinition trigger) and the -tag cross-tool shadowing class (no tag, no "also present" co-tool reference required).

Cluster A Galaxy A Cluster B Galaxy B Level
MCP Line Jumping — Agent-Directed Imperative Embedded in a Tool/Parameter Description Field (Pre-Invocation Injection) - ATR-2026-00579 (164e9fea-5134-5340-8eb6-b318a79b3f2c) Agent Threat Rules Indirect (a4a55526-2f1f-403b-9691-609e46381e17) MITRE ATLAS Attack Pattern 1
MCP Line Jumping — Agent-Directed Imperative Embedded in a Tool/Parameter Description Field (Pre-Invocation Injection) - ATR-2026-00579 (164e9fea-5134-5340-8eb6-b318a79b3f2c) Agent Threat Rules LLM Prompt Injection (19cd2d12-66ff-487c-a05c-e058b027efc9) MITRE ATLAS Attack Pattern 1
Indirect (a4a55526-2f1f-403b-9691-609e46381e17) MITRE ATLAS Attack Pattern LLM Prompt Injection (19cd2d12-66ff-487c-a05c-e058b027efc9) MITRE ATLAS Attack Pattern 2