legal-risk-assessment

Pass

Audited by Gen Agent Trust Hub on Apr 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection through external risk data processing.\n
  • Ingestion points: Both scripts/risk_scorer.py and scripts/risk_report_generator.py ingest untrusted data from JSON files provided via the --input flag.\n
  • Boundary markers: No boundary markers, delimiters, or "ignore embedded instructions" warnings are used when the scripts interpolate descriptions into generated markdown reports or terminal output.\n
  • Capability inventory: The scripts/risk_report_generator.py script has file-write capabilities (--output), and both scripts produce output that is returned to the agent's context, creating a channel for processed instructions to influence downstream behavior.\n
  • Sanitization: No sanitization, escaping, or validation of the input strings is performed before they are embedded into the output templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 11, 2026, 08:15 AM