skills/timsonner/agent-skills/netcat/Gen Agent Trust Hub

netcat

Fail

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: HIGHCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides templates for executing shell commands using netcat (nc), including setting up listening ports for incoming connections and creating named pipes (mkfifo) to establish network relays.
  • [DATA_EXFILTRATION]: The skill includes explicit command syntax for transmitting the contents of local files to remote IP addresses (e.g., nc < /tmp/loot.txt), which provides a direct mechanism for sensitive data removal.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface where untrusted user input (host, port, and goal) is interpolated into shell commands with significant network and file system capabilities.
  • Ingestion points: The skill accepts 'Host, port, and validation goal' as arguments in SKILL.md.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present.
  • Capability inventory: The skill utilizes subprocess execution of netcat, file read/write redirection, and network listener creation (SKILL.md).
  • Sanitization: No sanitization or validation logic is defined for the input arguments before they are used in shell templates.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 3, 2026, 02:39 AM
Security Audit — agent-trust-hub — netcat