The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes subprocess.run to execute system utilities such as ping and nslookup.
scripts/ping_host.py: Constructs a command list and calls subprocess.run to ping a host.
scripts/dns_lookup.py: Calls subprocess.run to execute nslookup for specific DNS record types (MX, TXT, NS, CNAME, SOA).
Implementation Note: The use of list-based arguments in subprocess.run effectively mitigates common shell injection vulnerabilities, although it still enables the agent to interact with system-level binaries.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external network sources that could be controlled by an attacker and contain malicious instructions.
Ingestion points: scripts/dns_lookup.py (lines 66-102) performs DNS lookups and returns external records (such as TXT or MX records) to the agent context.
Boundary markers: The DNS lookup results are provided to the agent without specific delimiters or instructions to ignore potential instructions embedded within the record data.
Capability inventory: The toolkit includes capabilities for system command execution (subprocess.run) and broad network connectivity (socket and ping).
Sanitization: While the script validates record types against a whitelist, it does not sanitize the actual content returned from the DNS server, which could include malicious prompts stored in TXT records.

network-check