news-stock-impact

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local Python script scripts/session_clock.py to resolve market session times and exchange holidays. This execution is limited to a bundled utility script using standard library modules for time-zone calculations.\n- [PROMPT_INJECTION]: The skill processes untrusted news headlines and search results, which presents an indirect prompt injection surface.\n
  • Ingestion points: Untrusted news data and headlines are fetched via live web tools as described in SKILL.md.\n
  • Boundary markers: The skill requires a strict output structure and multi-axis ranking system to ensure data is processed analytically.\n
  • Capability inventory: The skill possesses web access for news gathering and local script execution capabilities.\n
  • Sanitization: Instructions include explicit filters to verify if news is a surprise, changes cash flows, or has already been priced in by the market.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 11:01 AM
Security Audit — agent-trust-hub — news-stock-impact