The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automates the execution of security tools such as Nmap and Nessus through Python and shell scripts.
Evidence: security_scan.py and nmap_scan_template.sh are listed as resources for performing scans.
[INDIRECT_PROMPT_INJECTION]: There is a potential attack surface where the skill ingests data from external network targets which could contain malicious instructions designed to manipulate the agent.
Ingestion points: report_parser.py processes external data from Nmap XML and Nessus CSV reports.
Boundary markers: Not mentioned or enforced in the skill documentation.
Capability inventory: The skill has the ability to execute shell commands and security tools.
Sanitization: No sanitization or validation of the ingested report data is described.
[CREDENTIALS_UNSAFE]: The documentation encourages users to provide authentication credentials directly to the agent to enable testing of protected resources.
Evidence: The Best Practices section states users should "Provide necessary authentication credentials for testing protected resources."

performing-security-testing