cve-watchlist-action-recommendation-generator

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external CVE scan results (JSON/SARIF). Maliciously crafted text within vulnerability titles or descriptions in the source scan files could influence the agent's report generation or subsequent actions.
  • Ingestion points: The scripts scripts/parse_scan_results.py and scripts/calculate_risk_score.py read untrusted data from various security scan result files.
  • Boundary markers: The skill does not implement explicit delimiters or instructions to ignore embedded commands within the processed data.
  • Capability inventory: The skill focuses on data processing and report generation (markdown output); it does not exhibit dangerous capabilities such as network access or arbitrary command execution based on the ingested content.
  • Sanitization: There is no evidence of string sanitization or validation of fields like 'description' or 'title' before they are used to generate the final report.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:20 PM