MCP Tool Description Poisoning: Local Pattern Inspection
Tool-integrity analysis inspects reported tool descriptions for embedded instruction patterns.
MCP and tool security
These guides document actual 1-SEC implementation paths and source fixtures. Start by verifying the event producer and field meanings, then read the coverage limits before choosing an enforcement policy.
MCP Rug Pulls: Description Hash Changes and Reapproval
Tool-integrity state records a tool's metadata hash and reports later changes.
Agent Tool Payloads: Shell, URI, and Destructive SQL Indicators
Containment collects strings from structured tool events and classifies selected shell primitives, risky pipelines, chained destructive SQL, command substitution, and unsafe URI schemes.
MCP Cross-Server Tool Shadowing: Name Collision Signals
Tool-integrity state tracks the reported server for a known tool name.
MCP Remote Route Monitoring: Reported URLs and Hostname Checks
Agent-action handling looks for MCP-related routing hints, extracts HTTP URLs from selected reported fields, and emits unauthorized_mcp_routing for a hostname outside localhost, 127.0.0.1, ::1, and internal-mcp.svc.cluster.local.
Hidden Unicode Instructions: Instrumenting the LLM Input Scanner
LLM input analysis decodes implemented evasion layers, scans normalized and sometimes raw text, and includes a zero-width-character injection pattern.