The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py executes Nmap scans using the python-nmap library to perform service version detection and vulnerability script execution on user-defined targets.
[EXTERNAL_DOWNLOADS]: The skill fetches the Known Exploited Vulnerabilities (KEV) catalog from the official CISA website (cisa.gov). This is a well-known and trusted government source.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted data from external sources and network scans.
Ingestion points: The skill ingests untrusted service version strings and script output via Nmap in scripts/agent.py and scan data from Nessus/Qualys APIs in SKILL.md.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the scripts or prompt templates.
Capability inventory: The agent has the capability to write files (vuln_report.json) and execute subprocesses (Nmap) as seen in scripts/agent.py.
Sanitization: The ingested data is processed directly without sanitization or validation before being included in reports or used for prioritization logic.
[SAFE]: The implementation in scripts/agent.py and code examples in SKILL.md disable SSL certificate verification (verify=False) for Nessus and ServiceNow API requests. This is a common but discouraged practice that makes connections vulnerable to Man-in-the-Middle (MITM) attacks.

building-vulnerability-scanning-workflow