Remote code-injection attacks are one of the most frequently used attacking vectors in computer security. To detect and analyze injected code (often called shellcode), some researchers have proposed networklevel code emulators. A network-level code emulator can detect shellcode accurately and help analysts to understand the behavior of shellcode. We demonstrated that memory-scanning attacks can evade current emulators, and propose Yataglass, an elaborated network-level code emulator, that enables us to analyze shellcode that incorporates memory-scanning attacks. According to our experimental results, Yataglass successfully emulated and analyzed real shellcode into which we had manually incorporated memory-scanning attacks. Key words: Network-level code emulation, Code-injection attack, Memoryscanning attack, Intrusion detection, Intrusion analysis