The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the dig utility to perform DNS lookups (MX, TXT, CNAME).
Evidence: Subprocess calls are located in scripts/domain_health.py and scripts/verify_email.py.
Observation: Commands are invoked using list arguments without the shell=True flag, which prevents shell-based command injection from user-supplied domain names.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data fetched from external DNS records, which represents a potential surface for indirect prompt injection if a domain owner places malicious instructions in a TXT or SPF record.
Ingestion points: DNS record values fetched via dig in scripts/domain_health.py.
Boundary markers: None; the script output is returned directly to the agent's context.
Capability inventory: The skill can execute shell commands (dig) and initiate network connections (SMTP probes).
Sanitization: No specific content filtering is performed on the DNS records before they are analyzed by the agent.

domain-health