MCPS: Per-Message Cryptographic Signing for the Model Context Protocol

he Model Context Protocol (MCP), introduced by Anthropic in 2024, has become the dominant interface for connecting AI agents to external tools and data sources, with            
  ecosystem-wide adoption exceeding 57 million weekly package downloads. Despite this rapid adoption, MCP lacks native cryptographic security at the message level. This paper        
  presents MCPS (MCP Secure), a backward-compatible cryptographic signing layer that provides per-message ECDSA P-256 signatures, replay protection, tool integrity verification, and 
  a portable agent identity mechanism. We describe the protocol design, analyse the threat model, and demonstrate sub-2ms overhead per message. An open-source reference              
  implementation is available as a zero-dependency npm package.