breakout-trade-planner

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading stock screener results from local JSON files provided by the user. While the skill performs schema validation, it lacks explicit boundary markers to separate processed data from instructions to the AI.
  • Ingestion points: User-provided JSON data is loaded in scripts/plan_breakout_trades.py.
  • Boundary markers: None identified.
  • Capability inventory: The skill possesses file-writing capabilities for reports but has no network or subprocess execution features.
  • Sanitization: The script performs validation on the JSON schema and checks for required fields.
  • [EXTERNAL_DOWNLOADS]: The skill requires the pandas-market-calendars library, which is a standard financial tool for market session calculation, downloaded from the official Python Package Index.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:41 PM
Security Audit — agent-trust-hub — breakout-trade-planner