The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a wide range of commands to execute the nmap utility for network reconnaissance, port scanning, and vulnerability detection across all instruction steps.
[DYNAMIC_EXECUTION]: Step 5 includes instructions to generate and execute a Python 3 script at runtime to parse XML results from the nmap scan.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and parse data originating from external networks, such as service banners, hostnames, and protocol headers, which could contain malicious payloads intended to influence the agent's behavior.
Ingestion points: scan-results.xml (processed in Step 5), which contains data retrieved from external target hosts.
Boundary markers: Absent. The instructions do not define delimiters or provide guidance to ignore instructions that might be embedded in service banners or network metadata.
Capability inventory: Shell command execution (nmap) and file system interaction (writing scan results to files).
Sanitization: Absent. The provided Python parsing script directly prints service names, versions, and IP addresses extracted from the XML output without validation or escaping.

nmap-recon