vulnerability-management

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (MEDIUM): The skill is susceptible to Indirect Prompt Injection through the ingestion of untrusted external data.
  • Ingestion points: VulnerabilityScanner.import_csv() and VulnerabilityScanner.import_nessus_csv() as described in references/REFERENCE.md and SKILL.md.
  • Boundary markers: None documented. There are no instructions provided to the agent on how to distinguish between data and embedded instructions within the CSV files.
  • Capability inventory: The skill processes this data to generate risk scores, technical reports, and executive summaries (generate_report, generate_risk_report). While it does not perform direct system modifications, its output is intended to drive security decision-making.
  • Sanitization: No input validation or sanitization routines are mentioned for the fields imported from CSV files (e.g., 'description' or 'solution').
  • Risk: An attacker who can influence a vulnerability scan result could insert instructions into vulnerability descriptions to trick the agent into misrepresenting the severity of issues or hiding specific findings from remediation reports.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 01:15 PM