The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides multiple examples for executing the hydra command-line utility to perform authentication testing against SSH, RDP, HTTP POST forms, FTP, and SMB. These operations involve executing subprocesses that interact with network services.
[DATA_EXFILTRATION]: The instructions direct the agent to read from local files such as users.txt and passwords.txt and transmit authentication attempts to external network targets. This pattern involves the reading of potentially sensitive local credential data followed by network transmission.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process output from network targets (via Hydra responses) and capture evidence, creating a surface for indirect prompt injection from untrusted remote systems.
Ingestion points: Command output from hydra and local wordlist files (users.txt, passwords.txt) referenced in the SKILL.md examples.
Boundary markers: None. The skill does not provide delimiters or instructions to ignore or sanitize embedded commands within the tool output or wordlists.
Capability inventory: Execution of arbitrary network authentication commands via the hydra binary.
Sanitization: No validation or sanitization logic is described for the data ingested from target responses before it is used for evidence capture or analysis.

hydra