The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute a Python one-liner that reads a file from a user-provided path. This creates a risk of command injection if the input path string contains shell metacharacters designed to escape the command's quoting.
[COMMAND_EXECUTION]: The skill's primary function is the dynamic analysis of untrusted shellcode, which involves loading and running arbitrary binary code within a debugger. Executing such code on a host system is inherently risky, particularly as the chosen sacrificial process (timeout.exe) does not provide a robust security sandbox.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted binary data from external files. An attacker could craft shellcode with specific disassembly patterns, strings, or markers to subvert the AI's analysis logic or produce misleading reports.
Ingestion points: Step 2 of SKILL.md reads a shellcode file from a path provided by the user.
Boundary markers: There are no protective delimiters or instructions provided to separate the untrusted shellcode data from the agent's operational logic.
Capability inventory: The skill possesses the ability to execute shell commands, modify process memory, and write files to the local system.
Sanitization: No validation or sanitization of the shellcode's binary structure or content is performed prior to analysis.

shellcode-analyzer