The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions to execute networking and security tools including nmap, hydra, swaks, netcat (nc), openssl, and dig to interact with remote SMTP servers.
[DYNAMIC_EXECUTION]: The reference file references/smtp-commands.md contains Python scripts designed to be executed by the agent to perform email spoofing and DKIM signing during the testing process.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external network services (banners, SMTP responses), which creates a potential attack surface where a malicious server could influence the agent's behavior.
Ingestion points: Output from nc, openssl, nmap, and SMTP server responses (e.g., banners, EHLO extensions).
Boundary markers: Not explicitly defined in the provided files.
Capability inventory: The skill uses swaks, hydra, and custom Python scripts which perform network operations and can send data externally.
Sanitization: No explicit sanitization of server responses is documented before the agent processes them.
[EXTERNAL_DOWNLOADS]: The documentation mentions installing third-party Python packages checkdmarc and dkimpy via pip3, which are legitimate tools for email security analysis.

smtp-pentesting